modelId
stringlengths
5
139
author
stringlengths
2
42
last_modified
timestamp[us, tz=UTC]date
2020-02-15 11:33:14
2025-08-02 18:27:42
downloads
int64
0
223M
likes
int64
0
11.7k
library_name
stringclasses
549 values
tags
listlengths
1
4.05k
pipeline_tag
stringclasses
55 values
createdAt
timestamp[us, tz=UTC]date
2022-03-02 23:29:04
2025-08-02 18:24:50
card
stringlengths
11
1.01M
mradermacher/mistral-7b-ppo-clean-hermes-GGUF
mradermacher
2024-10-13T12:20:06Z
17
0
transformers
[ "transformers", "gguf", "en", "base_model:HINT-lab/mistral-7b-ppo-clean-hermes", "base_model:quantized:HINT-lab/mistral-7b-ppo-clean-hermes", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T11:43:09Z
--- base_model: HINT-lab/mistral-7b-ppo-clean-hermes language: - en library_name: transformers quantized_by: mradermacher tags: [] --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/HINT-lab/mistral-7b-ppo-clean-hermes <!-- provided-files --> weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q2_K.gguf) | Q2_K | 2.8 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q3_K_S.gguf) | Q3_K_S | 3.3 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q3_K_M.gguf) | Q3_K_M | 3.6 | lower quality | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q3_K_L.gguf) | Q3_K_L | 3.9 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.IQ4_XS.gguf) | IQ4_XS | 4.0 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q4_K_S.gguf) | Q4_K_S | 4.2 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q4_K_M.gguf) | Q4_K_M | 4.5 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q5_K_S.gguf) | Q5_K_S | 5.1 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q5_K_M.gguf) | Q5_K_M | 5.2 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q6_K.gguf) | Q6_K | 6.0 | very good quality | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.Q8_0.gguf) | Q8_0 | 7.8 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-clean-hermes-GGUF/resolve/main/mistral-7b-ppo-clean-hermes.f16.gguf) | f16 | 14.6 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
adipanda/deku-simpletuner-lora-1
adipanda
2024-10-13T12:19:06Z
5
0
diffusers
[ "diffusers", "flux", "flux-diffusers", "text-to-image", "simpletuner", "safe-for-work", "lora", "template:sd-lora", "lycoris", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "license:other", "region:us" ]
text-to-image
2024-10-12T04:23:30Z
--- license: other base_model: "black-forest-labs/FLUX.1-dev" tags: - flux - flux-diffusers - text-to-image - diffusers - simpletuner - safe-for-work - lora - template:sd-lora - lycoris inference: true widget: - text: 'unconditional (blank prompt)' parameters: negative_prompt: 'blurry, cropped, ugly' output: url: ./assets/image_0_0.png - text: 'A scene from My Hero Academia. Izuku Midoriya holding a sign that says ''I LOVE PROMPTS!'', he is standing full body on a beach at sunset. He is wearing a green hero costume with protective padding and a metallic collar. The setting sun casts a dynamic shadow on his face.' parameters: negative_prompt: 'blurry, cropped, ugly' output: url: ./assets/image_1_0.png - text: 'A scene from My Hero Academia. Izuku Midoriya jumping out of a propeller airplane, sky diving. He looks excited and his hair is blowing in the wind. The sky is clear and blue, there are birds pictured in the distance.' parameters: negative_prompt: 'blurry, cropped, ugly' output: url: ./assets/image_2_0.png - text: 'A scene from My Hero Academia. Izuku Midoriya spinning a basketball on his finger on a basketball court. He is wearing a lakers jersey with the #12 on it. The basketball hoop and crowd are in the background cheering him. He is smiling.' parameters: negative_prompt: 'blurry, cropped, ugly' output: url: ./assets/image_3_0.png - text: 'A scene from My Hero Academia. Izuku Midoriya is wearing a suit in an office shaking the hand of a business woman. The woman has purple hair and is wearing professional attire. There is a Google logo in the background. It is during daytime, and the overall sentiment is one of accomplishment.' parameters: negative_prompt: 'blurry, cropped, ugly' output: url: ./assets/image_4_0.png - text: 'A scene from My Hero Academia. Izuku Midoriya is fighting a large brown grizzly bear, deep in a forest. The bear is tall and standing on two legs, roaring. The bear is also wearing a crown because it is the king of all bears. Around them are tall trees and other animals watching.' parameters: negative_prompt: 'blurry, cropped, ugly' output: url: ./assets/image_5_0.png --- # deku-simpletuner-lora-1 This is a LyCORIS adapter derived from [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev). No validation prompt was used during training. None ## Validation settings - CFG: `3.5` - CFG Rescale: `0.0` - Steps: `20` - Sampler: `None` - Seed: `42` - Resolution: `1024x1024` Note: The validation settings are not necessarily the same as the [training settings](#training-settings). You can find some example images in the following gallery: <Gallery /> The text encoder **was not** trained. You may reuse the base model text encoder for inference. ## Training settings - Training epochs: 166 - Training steps: 3000 - Learning rate: 0.0003 - Effective batch size: 48 - Micro-batch size: 48 - Gradient accumulation steps: 1 - Number of GPUs: 1 - Prediction type: flow-matching - Rescaled betas zero SNR: False - Optimizer: adamw_bf16 - Precision: Pure BF16 - Quantised: Yes: int8-quanto - Xformers: Not used - LyCORIS Config: ```json { "algo": "lokr", "multiplier": 1.0, "linear_dim": 10000, "linear_alpha": 1, "factor": 12, "apply_preset": { "target_module": [ "Attention", "FeedForward" ], "module_algo_map": { "Attention": { "factor": 12 }, "FeedForward": { "factor": 6 } } } } ``` ## Datasets ### deku-512 - Repeats: 2 - Total number of images: 276 - Total number of aspect buckets: 1 - Resolution: 0.262144 megapixels - Cropped: False - Crop style: None - Crop aspect: None ## Inference ```python import torch from diffusers import DiffusionPipeline from lycoris import create_lycoris_from_weights model_id = 'black-forest-labs/FLUX.1-dev' adapter_id = 'pytorch_lora_weights.safetensors' # you will have to download this manually lora_scale = 1.0 wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_id, pipeline.transformer) wrapper.merge_to() prompt = "An astronaut is riding a horse through the jungles of Thailand." pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') image = pipeline( prompt=prompt, num_inference_steps=20, generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826), width=1024, height=1024, guidance_scale=3.5, ).images[0] image.save("output.png", format="PNG") ```
interview-eval/zephyr-7b-stem-case-5
interview-eval
2024-10-13T12:09:13Z
9
0
transformers
[ "transformers", "safetensors", "mistral", "text-generation", "alignment-handbook", "trl", "sft", "generated_from_trainer", "conversational", "dataset:scottsuk0306/DepthQA", "base_model:alignment-handbook/zephyr-7b-sft-full", "base_model:finetune:alignment-handbook/zephyr-7b-sft-full", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T11:59:05Z
--- library_name: transformers license: apache-2.0 base_model: alignment-handbook/zephyr-7b-sft-full tags: - alignment-handbook - trl - sft - generated_from_trainer - trl - sft - generated_from_trainer datasets: - scottsuk0306/DepthQA model-index: - name: zephyr-7b-stem-case-5 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # zephyr-7b-stem-case-5 This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the scottsuk0306/DepthQA dataset. It achieves the following results on the evaluation set: - Loss: 0.1790 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - distributed_type: multi-GPU - num_devices: 8 - total_train_batch_size: 64 - total_eval_batch_size: 64 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: cosine - lr_scheduler_warmup_ratio: 0.03 - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 1.1109 | 1.0 | 2 | 1.3244 | | 1.1109 | 2.0 | 4 | 0.9690 | | 1.1415 | 3.0 | 6 | 0.8217 | | 1.1415 | 4.0 | 8 | 0.6269 | | 0.7246 | 5.0 | 10 | 0.4879 | | 0.7246 | 6.0 | 12 | 0.3759 | | 0.7246 | 7.0 | 14 | 0.2801 | | 0.3832 | 8.0 | 16 | 0.2172 | | 0.3832 | 9.0 | 18 | 0.1864 | | 0.2054 | 10.0 | 20 | 0.1790 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.0 - Tokenizers 0.19.1
Zuntan/flux1-dev-TurboA-8step
Zuntan
2024-10-13T12:02:19Z
11
2
null
[ "gguf", "license:other", "region:us" ]
null
2024-10-13T09:24:19Z
--- license: other license_name: flux-1-dev-non-commercial-license license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md --- [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev) に [alimama-creative/FLUX.1-Turbo-Alpha](https://huggingface.co/alimama-creative/FLUX.1-Turbo-Alpha) をマージして、GGUF と nf4 に変換したモデルです。
cuongdev/haizzzz
cuongdev
2024-10-13T12:00:18Z
35
0
diffusers
[ "diffusers", "text-to-image", "stable-diffusion", "license:creativeml-openrail-m", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionPipeline", "region:us" ]
text-to-image
2024-10-13T11:57:02Z
--- license: creativeml-openrail-m tags: - text-to-image - stable-diffusion --- ### haizzzz Dreambooth model trained by cuongdev with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) Sample pictures of this concept:
interview-eval/zephyr-7b-math-case-6
interview-eval
2024-10-13T11:48:26Z
13
0
transformers
[ "transformers", "safetensors", "mistral", "text-generation", "alignment-handbook", "trl", "sft", "generated_from_trainer", "conversational", "dataset:EunsuKim/GSM8K", "dataset:EunsuKim/MATH", "base_model:alignment-handbook/zephyr-7b-sft-full", "base_model:finetune:alignment-handbook/zephyr-7b-sft-full", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T11:31:52Z
--- library_name: transformers license: apache-2.0 base_model: alignment-handbook/zephyr-7b-sft-full tags: - alignment-handbook - trl - sft - generated_from_trainer - trl - sft - generated_from_trainer datasets: - EunsuKim/GSM8K - EunsuKim/MATH model-index: - name: zephyr-7b-math-case-6 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # zephyr-7b-math-case-6 This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the EunsuKim/GSM8K and the EunsuKim/MATH datasets. It achieves the following results on the evaluation set: - Loss: 0.0198 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - distributed_type: multi-GPU - num_devices: 8 - total_train_batch_size: 64 - total_eval_batch_size: 64 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: cosine - lr_scheduler_warmup_ratio: 0.03 - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 0.8827 | 1.0 | 5 | 0.7429 | | 0.6709 | 2.0 | 10 | 0.5531 | | 0.5071 | 3.0 | 15 | 0.4035 | | 0.3519 | 4.0 | 20 | 0.2455 | | 0.2036 | 5.0 | 25 | 0.1302 | | 0.1035 | 6.0 | 30 | 0.0602 | | 0.0527 | 7.0 | 35 | 0.0356 | | 0.0321 | 8.0 | 40 | 0.0249 | | 0.0236 | 9.0 | 45 | 0.0206 | | 0.0202 | 10.0 | 50 | 0.0198 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.0 - Tokenizers 0.19.1
mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF
mradermacher
2024-10-13T11:41:51Z
94
0
transformers
[ "transformers", "gguf", "mergekit", "merge", "en", "base_model:BenevolenceMessiah/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0", "base_model:quantized:BenevolenceMessiah/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T09:43:01Z
--- base_model: BenevolenceMessiah/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0 language: - en library_name: transformers quantized_by: mradermacher tags: - mergekit - merge --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/BenevolenceMessiah/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0 <!-- provided-files --> weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-i1-GGUF ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q2_K.gguf) | Q2_K | 3.1 | | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q3_K_S.gguf) | Q3_K_S | 3.6 | | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q3_K_M.gguf) | Q3_K_M | 3.9 | lower quality | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q3_K_L.gguf) | Q3_K_L | 4.2 | | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.IQ4_XS.gguf) | IQ4_XS | 4.3 | | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q4_K_S.gguf) | Q4_K_S | 4.6 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q4_K_M.gguf) | Q4_K_M | 4.8 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q5_K_S.gguf) | Q5_K_S | 5.4 | | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q5_K_M.gguf) | Q5_K_M | 5.5 | | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q6_K.gguf) | Q6_K | 6.4 | very good quality | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.Q8_0.gguf) | Q8_0 | 8.2 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0-GGUF/resolve/main/Qwen2.5-Coder-7B-3x-Instruct-TIES-v1.0.f16.gguf) | f16 | 15.3 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
Malekhmem/mistralv3_malek
Malekhmem
2024-10-13T11:32:21Z
87
0
transformers
[ "transformers", "safetensors", "mistral", "text-generation", "trl", "sft", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "4-bit", "bitsandbytes", "region:us" ]
text-generation
2024-10-07T19:02:06Z
--- library_name: transformers tags: - trl - sft --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
steffygreypaul/Experiment8
steffygreypaul
2024-10-13T11:30:27Z
93
0
transformers
[ "transformers", "safetensors", "llama", "text-generation", "conversational", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T11:29:16Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
SkyWater21/mbert-lv-go-emotions-ekman
SkyWater21
2024-10-13T11:25:34Z
11
0
null
[ "safetensors", "bert", "lv", "dataset:SkyWater21/lv_go_emotions", "license:mit", "region:us" ]
null
2024-10-13T11:02:06Z
--- license: mit datasets: - SkyWater21/lv_go_emotions language: - lv --- Fine-tuned [multilingual BERT](https://huggingface.co/google-bert/bert-base-multilingual-cased) for multi-label emotion classification task. Model was trained on [lv_go_emotions](https://huggingface.co/datasets/SkyWater21/lv_go_emotions) dataset. This dataset is Latvian translation of [GoEmotions](https://huggingface.co/datasets/go_emotions) dataset. Google Translate was used to generate the machine translation. Original 26 emotions were mapped to 6 base emotions as per Dr. Ekman theory. Labels predicted by classifier: ```yaml 0: anger 1: disgust 2: fear 3: joy 4: sadness 5: surprise 6: neutral ``` Label mapping from 27 emotions from GoEmotion to 6 base emotions as per Dr. Ekman theory: |GoEmotion|Ekman| |---|---| | admiration | joy| | amusement | joy| | anger | anger| | annoyance | anger| | approval | joy| | caring | joy| | confusion | surprise| | curiosity | surprise| | desire | joy| | disappointment | sadness| | disapproval | anger| | disgust | disgust| | embarrassment | sadness| | excitement | joy| | fear | fear| | gratitude | joy| | grief | sadness| | joy | joy| | love | joy| | nervousness | fear| | optimism | joy| | pride | joy| | realization | surprise| | relief | joy| | remorse | sadness| | sadness | sadness| | surprise | surprise| | neutral | neutral| Seed used for random number generator is 42: ```python def set_seed(seed=42): random.seed(seed) np.random.seed(seed) torch.manual_seed(seed) if torch.cuda.is_available(): torch.cuda.manual_seed_all(seed) ``` Training parameters: ```yaml max_length: null batch_size: 64 shuffle: True num_workers: 8 pin_memory: False drop_last: False optimizer: adam lr: 0.00001 weight_decay: 0 problem_type: multi_label_classification num_epochs: 4 ``` Evaluation results on test split of [lv_go_emotions](https://huggingface.co/datasets/SkyWater21/lv_go_emotions/viewer/simplified_ekman) | |Precision|Recall|F1-Score|AUC-ROC|Support| |--------------|---------|------|--------|-------|-------| |anger | 0.58| 0.36| 0.45| 0.83| 726| |disgust | 0.88| 0.12| 0.21| 0.90| 123| |fear | 0.75| 0.48| 0.58| 0.93| 98| |joy | 0.82| 0.76| 0.79| 0.90| 2104| |sadness | 0.69| 0.46| 0.55| 0.88| 379| |surprise | 0.61| 0.51| 0.55| 0.87| 677| |neutral | 0.65| 0.62| 0.64| 0.83| 1787| |micro avg | 0.71| 0.60| 0.65| 0.92| 5894| |macro avg | 0.71| 0.47| 0.54| 0.88| 5894| |weighted avg | 0.71| 0.60| 0.64| 0.87| 5894| |samples avg | 0.63| 0.62| 0.62| nan| 5894|
physthoth/qua_llama_3_8B
physthoth
2024-10-13T11:24:03Z
77
0
transformers
[ "transformers", "safetensors", "llama", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "4-bit", "bitsandbytes", "region:us" ]
text-generation
2024-10-13T11:19:52Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
mav23/Higgs-Llama-3-70B-GGUF
mav23
2024-10-13T11:20:19Z
62
0
null
[ "gguf", "base_model:meta-llama/Meta-Llama-3-70B", "base_model:quantized:meta-llama/Meta-Llama-3-70B", "license:other", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T05:20:17Z
--- license: other base_model: meta-llama/Meta-Llama-3-70B --- # Higgs-Llama-3-70B Higgs-Llama-3-70B is post-trained from [meta-llama/Meta-Llama-3-70B](https://huggingface.co/meta-llama/Meta-Llama-3-70B), specially tuned for role-playing while being competitive in general-domain instruction-following and reasoning. We perform supervised fine-tuning with our in-house instruction-following and chat datasets. Afterwards, we construct preference pairs with a semi-automated pipeline that relies on both human-labelers and our private LLMs. We conduct iterative preference optimization to align the model. During alignment, we adopted a special strategy to align the model’s behavior with the system message. Compared with other instruct models, Higgs models follow their roles more closely. See our [release blog](https://boson.ai/higgs-opensource/). ## Evaluation All benchmarks lead to eventual overfitting, including those for LLMs. Training on data, particularly beneficial for benchmarks typically does not improve (or even worsen) role-playing performance. We worked to exclude benchmark data, including their training examples, from our fine-tuning data. We highlight our results on two new and challenging benchmarks: [MMLU-Pro](https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro) and [Arena-Hard](https://github.com/lm-sys/arena-hard-auto). MMLU-Pro extends the popular MMLU benchmark. We believe that it suffers from less overfitting by other released models as well, as it was released only recently (it was released after our models finished training). ### MMLU-Pro <table class="col-12 col-md-6" width="100px"> <tr> <td><b>Model</b></td> <td><b>MMLU-Pro</b></td> </tr> <tr> <td>GPT-4o</td> <td>72.6</td> </tr> <tr> <td>Gemini-1.5-Pro</td> <td>69.0</td> </tr> <tr> <td>Claude-3-Opus</td> <td>68.5</td> </tr> <tr> <td>GPT-4-Turbo</td> <td>63.7</td> </tr> <tr style="font-weight: bold"> <td>Higgs-Llama-3-70B</td> <td>63.2</td> </tr> <tr> <td>Gemini-1.5-Flash</td> <td>59.1</td> </tr> <tr> <td>Claude-3-Sonnet</td> <td>56.8</td> </tr> <tr> <td>Llama-3-70B-Instruct</td> <td>56.2</td> </tr> </table> ### Arena-Hard <table class="col-12 col-md-6"> <tr> <td><b>Model</b></td> <td><b>Arena-Hard</b></td> </tr> <tr> <td>GPT-4o</td> <td>79.5</td> </tr> <tr> <td>Gemini-1.5-Pro</td> <td>72.0</td> </tr> <tr> <td>Claude-3-Opus</td> <td>60.4</td> </tr> <tr style="font-weight: bold"> <td>Higgs-Llama-3-70B</td> <td>49.6</td> </tr> <tr> <td>Gemini-1.5-Flash</td> <td>49.6</td> </tr> <tr> <td>Claude-3-Sonnet</td> <td>46.8</td> </tr> <tr> <td>Claude-3-Haiku</td> <td>41.5</td> </tr> <tr> <td>Llama-3-70B-Instruct</td> <td>41.1</td> </tr> <tr> <td>GPT-4-0613</td> <td>37.9</td> </tr> <tr> <td>Mistral-Large</td> <td>37.7</td> </tr> </table> ## Overall Results In the following, we compare our model's performance with `gpt-4o` and `Llama-3-70B-Instruct` on [MMLU-Pro](https://github.com/TIGER-AI-Lab/MMLU-Pro), [Arena-Hard](https://github.com/lm-sys/arena-hard-auto/tree/main), [AlpacaEval 2.0 LC](https://github.com/tatsu-lab/alpaca_eval), MMLU, GPQA and DROP. For MMLU, GPQA and DROP, we adopt [openai/simple-evals](https://github.com/openai/simple-evals) for evaluation. For the other benchmarks, we evaluate via the official implementation. <div style="overflow: auto"> <table> <tr> <th></th> <td><b>MMLU-Pro</td> <td><b>Arena-Hard</td> <td><b>AlpacaEval <br> 2.0 LC</b></td> <td><b>MMLU</b></td> <td><b>GPQA</b></td> <td><b>DROP <br> (F1,3-shot)</b></td> </tr> <tr> <td>GPT-4o</td> <td>72.6</td> <td>79.5*</td> <td>57.5</td> <td>87.2</td> <td>49.9</td> <td>83.7</td> </tr> <tr style="font-weight: bold"> <td>Higgs-Llama-3-70B</td> <td>63.2</td> <td>49.6</td> <td>38.6</td> <td>80.8</td> <td>42.1</td> <td>81.6</td> </tr> <tr> <td>Llama-3-70B-Instruct*</td> <td>56.2</td> <td>41.1</td> <td>34.4</td> <td>80.2</td> <td>41.3</td> <td>81.4</td> </tr> </table> </div> <small>*For Llama-3-70B-Instruct, the MMLU-Pro number is copied from the [MMLU-Pro leaderboard](https://huggingface.co/spaces/TIGER-Lab/MMLU-Pro); the Arena-Hard numbers are copied from the [leaderboard updated on 5/21](https://github.com/lm-sys/arena-hard-auto/tree/main?tab=readme-ov-file#full-leaderboard-updated-0521) while we run gpt-4o ourselves; and the MMLU/GPQA/DROP are copied from [simple-evals](https://github.com/openai/simple-evals).</small> ## How to use We use the same prompting format as in Meta-Llama-3-70B-Instruct. ### Use with transformers See the snippet below for usage with Transformers: ```python import transformers import torch model_id = "bosonai/Higgs-Llama-3-70B" pipeline = transformers.pipeline( "text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device_map="auto", ) messages = [ {"role": "system", "content": "You are an AI assistant that speaks in the style of Sheldon Cooper. You are arguing with the user and is trying to prove the opposite of what the user said."}, {"role": "user", "content": "The earth is round."}, ] prompt = pipeline.tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True ) outputs = pipeline( prompt, max_new_tokens=256, eos_token_id=[ pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>"), pipeline.tokenizer.eos_token_id, ], do_sample=True, temperature=1.0, top_p=0.95, ) print(outputs[0]["generated_text"][len(prompt):]) ``` ## License [Our license](https://huggingface.co/bosonai/Higgs-Llama-3-70B/blob/main/LICENSE) is based on Meta's LLama 3 Community License.
RichardErkhov/tiiuae_-_falcon-7b-gguf
RichardErkhov
2024-10-13T11:18:12Z
6
0
null
[ "gguf", "arxiv:2205.14135", "arxiv:1911.02150", "arxiv:2101.00027", "arxiv:2005.14165", "arxiv:2104.09864", "arxiv:2306.01116", "endpoints_compatible", "region:us" ]
null
2024-10-13T09:58:39Z
Quantization made by Richard Erkhov. [Github](https://github.com/RichardErkhov) [Discord](https://discord.gg/pvy7H8DZMG) [Request more models](https://github.com/RichardErkhov/quant_request) falcon-7b - GGUF - Model creator: https://huggingface.co/tiiuae/ - Original model: https://huggingface.co/tiiuae/falcon-7b/ | Name | Quant method | Size | | ---- | ---- | ---- | | [falcon-7b.Q2_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q2_K.gguf) | Q2_K | 3.59GB | | [falcon-7b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.IQ3_XS.gguf) | IQ3_XS | 3.59GB | | [falcon-7b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.IQ3_S.gguf) | IQ3_S | 3.59GB | | [falcon-7b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q3_K_S.gguf) | Q3_K_S | 3.59GB | | [falcon-7b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.IQ3_M.gguf) | IQ3_M | 3.71GB | | [falcon-7b.Q3_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q3_K.gguf) | Q3_K | 3.86GB | | [falcon-7b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q3_K_M.gguf) | Q3_K_M | 3.86GB | | [falcon-7b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q3_K_L.gguf) | Q3_K_L | 4.08GB | | [falcon-7b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.IQ4_XS.gguf) | IQ4_XS | 3.89GB | | [falcon-7b.Q4_0.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q4_0.gguf) | Q4_0 | 3.92GB | | [falcon-7b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.IQ4_NL.gguf) | IQ4_NL | 3.96GB | | [falcon-7b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q4_K_S.gguf) | Q4_K_S | 4.42GB | | [falcon-7b.Q4_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q4_K.gguf) | Q4_K | 4.63GB | | [falcon-7b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q4_K_M.gguf) | Q4_K_M | 4.63GB | | [falcon-7b.Q4_1.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q4_1.gguf) | Q4_1 | 4.32GB | | [falcon-7b.Q5_0.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q5_0.gguf) | Q5_0 | 4.73GB | | [falcon-7b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q5_K_S.gguf) | Q5_K_S | 4.98GB | | [falcon-7b.Q5_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q5_K.gguf) | Q5_K | 5.34GB | | [falcon-7b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q5_K_M.gguf) | Q5_K_M | 5.34GB | | [falcon-7b.Q5_1.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q5_1.gguf) | Q5_1 | 5.13GB | | [falcon-7b.Q6_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q6_K.gguf) | Q6_K | 6.55GB | | [falcon-7b.Q8_0.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-gguf/blob/main/falcon-7b.Q8_0.gguf) | Q8_0 | 7.14GB | Original model description: --- datasets: - tiiuae/falcon-refinedweb language: - en inference: false license: apache-2.0 new_version: tiiuae/falcon-11B --- # 🚀 Falcon-7B **Falcon-7B is a 7B parameters causal decoder-only model built by [TII](https://www.tii.ae) and trained on 1,500B tokens of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) enhanced with curated corpora. It is made available under the Apache 2.0 license.** *Paper coming soon* 😊. 🤗 To get started with Falcon (inference, finetuning, quantization, etc.), we recommend reading [this great blogpost fron HF](https://huggingface.co/blog/falcon)! ## Why use Falcon-7B? * **It outperforms comparable open-source models** (e.g., [MPT-7B](https://huggingface.co/mosaicml/mpt-7b), [StableLM](https://github.com/Stability-AI/StableLM), [RedPajama](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-7B-v0.1) etc.), thanks to being trained on 1,500B tokens of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) enhanced with curated corpora. See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard). * **It features an architecture optimized for inference**, with FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135)) and multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)). * **It is made available under a permissive Apache 2.0 license allowing for commercial use**, without any royalties or restrictions. ⚠️ **This is a raw, pretrained model, which should be further finetuned for most usecases.** If you are looking for a version better suited to taking generic instructions in a chat format, we recommend taking a look at [Falcon-7B-Instruct](https://huggingface.co/tiiuae/falcon-7b-instruct). 🔥 **Looking for an even more powerful model?** [Falcon-40B](https://huggingface.co/tiiuae/falcon-40b) is Falcon-7B's big brother! ```python from transformers import AutoTokenizer, AutoModelForCausalLM import transformers import torch model = "tiiuae/falcon-7b" tokenizer = AutoTokenizer.from_pretrained(model) pipeline = transformers.pipeline( "text-generation", model=model, tokenizer=tokenizer, torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto", ) sequences = pipeline( "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:", max_length=200, do_sample=True, top_k=10, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id, ) for seq in sequences: print(f"Result: {seq['generated_text']}") ``` 💥 **Falcon LLMs require PyTorch 2.0 for use with `transformers`!** For fast inference with Falcon, check-out [Text Generation Inference](https://github.com/huggingface/text-generation-inference)! Read more in this [blogpost]((https://huggingface.co/blog/falcon). You will need **at least 16GB of memory** to swiftly run inference with Falcon-7B. # Model Card for Falcon-7B ## Model Details ### Model Description - **Developed by:** [https://www.tii.ae](https://www.tii.ae); - **Model type:** Causal decoder-only; - **Language(s) (NLP):** English, German, Spanish, French (and limited capabilities in Italian, Portuguese, Polish, Dutch, Romanian, Czech, Swedish); - **License:** Apache 2.0. ### Model Source - **Paper:** *coming soon*. ## Uses ### Direct Use Research on large language models; as a foundation for further specialization and finetuning for specific usecases (e.g., summarization, text generation, chatbot, etc.) ### Out-of-Scope Use Production use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful. ## Bias, Risks, and Limitations Falcon-7B is trained on English and French data only, and will not generalize appropriately to other languages. Furthermore, as it is trained on a large-scale corpora representative of the web, it will carry the stereotypes and biases commonly encountered online. ### Recommendations We recommend users of Falcon-7B to consider finetuning it for the specific set of tasks of interest, and for guardrails and appropriate precautions to be taken for any production use. ## How to Get Started with the Model ```python from transformers import AutoTokenizer, AutoModelForCausalLM import transformers import torch model = "tiiuae/falcon-7b" tokenizer = AutoTokenizer.from_pretrained(model) pipeline = transformers.pipeline( "text-generation", model=model, tokenizer=tokenizer, torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto", ) sequences = pipeline( "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:", max_length=200, do_sample=True, top_k=10, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id, ) for seq in sequences: print(f"Result: {seq['generated_text']}") ``` ## Training Details ### Training Data Falcon-7B was trained on 1,500B tokens of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb), a high-quality filtered and deduplicated web dataset which we enhanced with curated corpora. Significant components from our curated copora were inspired by The Pile ([Gao et al., 2020](https://arxiv.org/abs/2101.00027)). | **Data source** | **Fraction** | **Tokens** | **Sources** | |--------------------|--------------|------------|-----------------------------------| | [RefinedWeb-English](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) | 79% | 1,185B | massive web crawl | | Books | 7% | 110B | | | Conversations | 6% | 85B | Reddit, StackOverflow, HackerNews | | Code | 3% | 45B | | | RefinedWeb-French | 3% | 45B | massive web crawl | | Technical | 2% | 30B | arXiv, PubMed, USPTO, etc. | The data was tokenized with the Falcon-[7B](https://huggingface.co/tiiuae/falcon-7b)/[40B](https://huggingface.co/tiiuae/falcon-40b) tokenizer. ### Training Procedure Falcon-7B was trained on 384 A100 40GB GPUs, using a 2D parallelism strategy (PP=2, DP=192) combined with ZeRO. #### Training Hyperparameters | **Hyperparameter** | **Value** | **Comment** | |--------------------|------------|-------------------------------------------| | Precision | `bfloat16` | | | Optimizer | AdamW | | | Learning rate | 6e-4 | 4B tokens warm-up, cosine decay to 1.2e-5 | | Weight decay | 1e-1 | | | Z-loss | 1e-4 | | | Batch size | 2304 | 30B tokens ramp-up | #### Speeds, Sizes, Times Training happened in early March 2023 and took about two weeks. ## Evaluation *Paper coming soon*. See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) for early results. ## Technical Specifications ### Model Architecture and Objective Falcon-7B is a causal decoder-only model trained on a causal language modeling task (i.e., predict the next token). The architecture is broadly adapted from the GPT-3 paper ([Brown et al., 2020](https://arxiv.org/abs/2005.14165)), with the following differences: * **Positionnal embeddings:** rotary ([Su et al., 2021](https://arxiv.org/abs/2104.09864)); * **Attention:** multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)) and FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135)); * **Decoder-block:** parallel attention/MLP with a single layer norm. | **Hyperparameter** | **Value** | **Comment** | |--------------------|-----------|----------------------------------------| | Layers | 32 | | | `d_model` | 4544 | Increased to compensate for multiquery | | `head_dim` | 64 | Reduced to optimise for FlashAttention | | Vocabulary | 65024 | | | Sequence length | 2048 | | ### Compute Infrastructure #### Hardware Falcon-7B was trained on AWS SageMaker, on 384 A100 40GB GPUs in P4d instances. #### Software Falcon-7B was trained a custom distributed training codebase, Gigatron. It uses a 3D parallelism approach combined with ZeRO and high-performance Triton kernels (FlashAttention, etc.) ## Citation *Paper coming soon* 😊. In the meanwhile, you can use the following information to cite: ``` @article{falcon40b, title={{Falcon-40B}: an open large language model with state-of-the-art performance}, author={Almazrouei, Ebtesam and Alobeidli, Hamza and Alshamsi, Abdulaziz and Cappelli, Alessandro and Cojocaru, Ruxandra and Debbah, Merouane and Goffinet, Etienne and Heslow, Daniel and Launay, Julien and Malartic, Quentin and Noune, Badreddine and Pannier, Baptiste and Penedo, Guilherme}, year={2023} } ``` To learn more about the pretraining dataset, see the 📓 [RefinedWeb paper](https://arxiv.org/abs/2306.01116). ``` @article{refinedweb, title={The {R}efined{W}eb dataset for {F}alcon {LLM}: outperforming curated corpora with web data, and web data only}, author={Guilherme Penedo and Quentin Malartic and Daniel Hesslow and Ruxandra Cojocaru and Alessandro Cappelli and Hamza Alobeidli and Baptiste Pannier and Ebtesam Almazrouei and Julien Launay}, journal={arXiv preprint arXiv:2306.01116}, eprint={2306.01116}, eprinttype = {arXiv}, url={https://arxiv.org/abs/2306.01116}, year={2023} } ``` ## License Falcon-7B is made available under the Apache 2.0 license. ## Contact [email protected]
mradermacher/mistral-7b-ppo-c-hermes-GGUF
mradermacher
2024-10-13T11:15:10Z
9
0
transformers
[ "transformers", "gguf", "en", "base_model:HINT-lab/mistral-7b-ppo-c-hermes", "base_model:quantized:HINT-lab/mistral-7b-ppo-c-hermes", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T10:46:48Z
--- base_model: HINT-lab/mistral-7b-ppo-c-hermes language: - en library_name: transformers quantized_by: mradermacher tags: [] --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/HINT-lab/mistral-7b-ppo-c-hermes <!-- provided-files --> weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q2_K.gguf) | Q2_K | 2.8 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q3_K_S.gguf) | Q3_K_S | 3.3 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q3_K_M.gguf) | Q3_K_M | 3.6 | lower quality | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q3_K_L.gguf) | Q3_K_L | 3.9 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.IQ4_XS.gguf) | IQ4_XS | 4.0 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q4_K_S.gguf) | Q4_K_S | 4.2 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q4_K_M.gguf) | Q4_K_M | 4.5 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q5_K_S.gguf) | Q5_K_S | 5.1 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q5_K_M.gguf) | Q5_K_M | 5.2 | | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q6_K.gguf) | Q6_K | 6.0 | very good quality | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.Q8_0.gguf) | Q8_0 | 7.8 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/mistral-7b-ppo-c-hermes-GGUF/resolve/main/mistral-7b-ppo-c-hermes.f16.gguf) | f16 | 14.6 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
b09501048/rtdetr4
b09501048
2024-10-13T11:06:54Z
119
0
transformers
[ "transformers", "safetensors", "rt_detr", "object-detection", "generated_from_trainer", "endpoints_compatible", "region:us" ]
object-detection
2024-10-13T11:06:47Z
--- library_name: transformers tags: - generated_from_trainer model-index: - name: rtdetr4 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # rtdetr4 This model was trained from scratch on an unknown dataset. ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3.0 ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.19.1
Triangle104/falcon-mamba-7b-Q5_K_S-GGUF
Triangle104
2024-10-13T10:45:09Z
10
0
null
[ "gguf", "llama-cpp", "gguf-my-repo", "en", "dataset:tiiuae/falcon-refinedweb", "dataset:HuggingFaceFW/fineweb-edu", "arxiv:2410.05355", "base_model:tiiuae/falcon-mamba-7b", "base_model:quantized:tiiuae/falcon-mamba-7b", "license:other", "model-index", "endpoints_compatible", "region:us" ]
null
2024-10-13T10:42:51Z
--- base_model: tiiuae/falcon-mamba-7b datasets: - tiiuae/falcon-refinedweb - HuggingFaceFW/fineweb-edu language: - en license: other license_name: falcon-mamba-7b-license license_link: https://falconllm.tii.ae/falcon-mamba-7b-terms-and-conditions.html tags: - llama-cpp - gguf-my-repo model-index: - name: falcon-mamba-7b results: - task: type: text-generation name: Text Generation dataset: name: IFEval (0-Shot) type: HuggingFaceH4/ifeval args: num_few_shot: 0 metrics: - type: inst_level_strict_acc and prompt_level_strict_acc value: 33.36 name: strict accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=tiiuae/falcon-mamba-7b name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: BBH (3-Shot) type: BBH args: num_few_shot: 3 metrics: - type: acc_norm value: 19.88 name: normalized accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=tiiuae/falcon-mamba-7b name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MATH Lvl 5 (4-Shot) type: hendrycks/competition_math args: num_few_shot: 4 metrics: - type: exact_match value: 3.63 name: exact match source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=tiiuae/falcon-mamba-7b name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: GPQA (0-shot) type: Idavidrein/gpqa args: num_few_shot: 0 metrics: - type: acc_norm value: 8.05 name: acc_norm source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=tiiuae/falcon-mamba-7b name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MuSR (0-shot) type: TAUR-Lab/MuSR args: num_few_shot: 0 metrics: - type: acc_norm value: 10.86 name: acc_norm source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=tiiuae/falcon-mamba-7b name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MMLU-PRO (5-shot) type: TIGER-Lab/MMLU-Pro config: main split: test args: num_few_shot: 5 metrics: - type: acc value: 14.47 name: accuracy source: url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=tiiuae/falcon-mamba-7b name: Open LLM Leaderboard --- # Triangle104/falcon-mamba-7b-Q5_K_S-GGUF This model was converted to GGUF format from [`tiiuae/falcon-mamba-7b`](https://huggingface.co/tiiuae/falcon-mamba-7b) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space. Refer to the [original model card](https://huggingface.co/tiiuae/falcon-mamba-7b) for more details on the model. --- Model Description Developed by: https://www.tii.ae Model type: Causal decoder-only Architecture: Mamba Language(s) (NLP): Mainly English License: TII Falcon-Mamba License 2.0 Usage Find below some example scripts on how to use the model in transformers (Make sure to have the latest transformers, or the one built from source): Using the Pytorch model Running the model on a CPU Click to expand Running the model on a GPU Click to expand Running the model on a GPU using torch.compile Click to expand Running the model on a GPU using different precisions FP16 Click to expand 4-bit Click to expand Training Details Training Data Falcon-Mamba has been trained with ~ 5,500 GT mainly coming from Refined-Web, a large volume web-only dataset filtered and deduplicated. Similar to the others Falcon suite models, Falcon-Mamba has been trained leveraging a multi-stage training strategy to increase the context-length from 2,048 to 8,192. Moreover, inspired by the concept of Curriculum Learning, we carefully selected data mixtures throughout the training stages, considering both data diversity and complexity. Note that at inference the context-length is not relevant as the Mamba architecture has no limit on long range dependency. At the last training stage, small portion of high-quality curated data was used to further enhance performance. Overall, the data sources included RefinedWeb-English, high quality technical data, code data and math data extracted from public sources. In particular, we used samples coming from Fineweb-edu during our last training stage. The data was tokenized with the Falcon-7B/11B tokenizer. Training Procedure Falcon-Mamba-7B was trained on 256 H100 80GB GPUs for the majority of the training, using a 3D parallelism strategy (TP=1, PP=1, DP=256) combined with ZeRO. Training Hyperparameters Hyperparameter Value Comment Precision bfloat16 Optimizer AdamW Max learning rate 6.4e-4 Following a WSD (warmup-stable-decay) learning rate schedule Weight decay 1e-1 Batch size 2048 The model was trained AdamW optimizer, WSD (warmup-stable-decay) learning rate schedule, and a batch size rampup from bmin=128b_{\mathrm{min}}=128bmin​=128 to bmax=2048b_{\mathrm{max}}=2048bmax​=2048 during first 50 GT of training. In the stable phase we used maximal learning rate ηmax=6.4×10−4\eta_{\mathrm{max}}=6.4 \times 10^{-4}ηmax​=6.4×10−4, and decayed it to the minimal value ηmin=ηmax256\eta_{\mathrm{min}}=\frac{\eta_{\mathrm{max}}}{256}ηmin​=256ηmax​​ with exponential schedule over 500 GT. Also, we applied BatchScaling during the rampup — rescaling learning rate η\etaη so that the Adam noise temperature Tnoise≡ηbT_{\mathrm{noise}}\equiv\frac{\eta}{\sqrt{b}}Tnoise​≡b ​η​ is kept constant. Speeds, Sizes, Times The model training took roughly two months. Evaluation Benchmarks We evaluate our model on all benchmarks of the new leaderboard's version using the lm-evaluation-harness package, and then normalize the evaluation results with HuggingFace score normalization. model name IFEval BBH MATH LvL5 GPQA MUSR MMLU-PRO Average Pure SSM models FalconMamba-7B 33.36 19.88 3.63 8.05 10.86 14.47 15.04 TRI-ML/mamba-7b-rw* 22.46 6.71 0.45 1.12 5.51 1.69 6.25 Hybrid SSM-attention models recurrentgemma-9b 30.76 14.80 4.83 4.70 6.60 17.88 13.20 Zyphra/Zamba-7B-v1* 24.06 21.12 3.32 3.03 7.74 16.02 12.55 Transformer models Falcon2-11B 32.61 21.94 2.34 2.80 7.53 15.44 13.78 Meta-Llama-3-8B 14.55 24.50 3.25 7.38 6.24 24.55 13.41 Meta-Llama-3.1-8B 12.70 25.29 4.61 6.15 8.98 24.95 13.78 Mistral-7B-v0.1 23.86 22.02 2.49 5.59 10.68 22.36 14.50 Mistral-Nemo-Base-2407 (12B) 16.83 29.37 4.98 5.82 6.52 27.46 15.08 gemma-7B 26.59 21.12 6.42 4.92 10.98 21.64 15.28 RWKV models RWKV-v6-Finch-7B* 27.65 9.04 1.11 2.81 2.25 5.85 8.12 RWKV-v6-Finch-14B* 29.81 12.89 1.13 5.01 3.16 11.3 10.55 Also, we evaluate our model on the benchmarks of the first leaderboard using lighteval. model name ARC HellaSwag MMLU Winogrande TruthfulQA GSM8K Average Pure SSM models FalconMamba-7B* 62.03 80.82 62.11 73.64 53.42 52.54 64.09 TRI-ML/mamba-7b-rw* 51.25 80.85 33.41 71.11 32.08 4.70 45.52 Hybrid SSM-attention models recurrentgemma-9b** 52.00 80.40 60.50 73.60 38.60 42.60 57.95 Zyphra/Zamba-7B-v1* 56.14 82.23 58.11 79.87 52.88 30.78 60.00 Transformer models Falcon2-11B 59.73 82.91 58.37 78.30 52.56 53.83 64.28 Meta-Llama-3-8B 60.24 82.23 66.70 78.45 42.93 45.19 62.62 Meta-Llama-3.1-8B 58.53 82.13 66.43 74.35 44.29 47.92 62.28 Mistral-7B-v0.1 59.98 83.31 64.16 78.37 42.15 37.83 60.97 Mistral-Nemo-Base-2407 (12B)* 57.94 82.82 64.43 73.72 49.14 55.27 63.89 gemma-7B 61.09 82.20 64.56 79.01 44.79 50.87 63.75 RWKV models RWKV-v6-Finch-7B* 43.86 75.19 41.69 68.27 42.19 19.64 48.47 RWKV-v6-Finch-14B* 47.44 78.86 52.33 71.27 45.45 38.06 55.57 Mostly, we took evaluation results from both leaderboards. For the models marked by star we evaluated the tasks internally, while for the models marked by two stars the results were taken from paper or model card. Throughput This model can achieve comparable throughput and performance compared to other transformer based models that use optimized kernels such as Flash Attention 2. Make sure to install the optimized Mamba kernels with the following commands: pip install "causal-conv1d>=1.4.0" mamba-ssm Refer to our FalconMamba blogpost for more details about performance evaluation. Technical Specifications Model Architecture and Objective Falcon-Mamba-7B is a causal decoder-only model trained on a causal language modeling task (i.e., predict the next token). The model is based on the Mamba architecture (Gu et al., 2023). Hyperparameter Value Comment Layers 64 Number of layers d_model 4096 Hidden dimension d_state 16 The SSM state dimension Vocabulary 65024 Vocabulary Size Sequence length 8192 During the last training stages Compute Infrastructure Hardware Falcon-Mamba-7B was trained on AWS SageMaker, using on average 256 H100 80GB GPUs in 32 p5 instances. Software Falcon-Mamba-7B was trained on an internal distributed training codebase, Gigatron. It uses a 3D parallelism approach combined with ZeRO, high-performance Triton kernels. Citation You can use the following bibtex citation: @misc{zuo2024falconmambacompetitiveattentionfree, title={Falcon Mamba: The First Competitive Attention-free 7B Language Model}, author={Jingwei Zuo and Maksim Velikanov and Dhia Eddine Rhaiem and Ilyas Chahed and Younes Belkada and Guillaume Kunsch and Hakim Hacid}, year={2024}, eprint={2410.05355}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2410.05355}, } Open LLM Leaderboard Evaluation Results Detailed results can be found here Metric Value Avg. 15.04 IFEval (0-Shot) 33.36 BBH (3-Shot) 19.88 MATH Lvl 5 (4-Shot) 3.63 GPQA (0-shot) 8.05 MuSR (0-shot) 10.86 MMLU-PRO (5-shot) 14.47 --- ## Use with llama.cpp Install llama.cpp through brew (works on Mac and Linux) ```bash brew install llama.cpp ``` Invoke the llama.cpp server or the CLI. ### CLI: ```bash llama-cli --hf-repo Triangle104/falcon-mamba-7b-Q5_K_S-GGUF --hf-file falcon-mamba-7b-q5_k_s.gguf -p "The meaning to life and the universe is" ``` ### Server: ```bash llama-server --hf-repo Triangle104/falcon-mamba-7b-Q5_K_S-GGUF --hf-file falcon-mamba-7b-q5_k_s.gguf -c 2048 ``` Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well. Step 1: Clone llama.cpp from GitHub. ``` git clone https://github.com/ggerganov/llama.cpp ``` Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux). ``` cd llama.cpp && LLAMA_CURL=1 make ``` Step 3: Run inference through the main binary. ``` ./llama-cli --hf-repo Triangle104/falcon-mamba-7b-Q5_K_S-GGUF --hf-file falcon-mamba-7b-q5_k_s.gguf -p "The meaning to life and the universe is" ``` or ``` ./llama-server --hf-repo Triangle104/falcon-mamba-7b-Q5_K_S-GGUF --hf-file falcon-mamba-7b-q5_k_s.gguf -c 2048 ```
kyledam/girllt
kyledam
2024-10-13T10:44:12Z
47
0
diffusers
[ "diffusers", "text-to-image", "flux", "lora", "template:sd-lora", "fluxgym", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "license:other", "region:us" ]
text-to-image
2024-10-13T07:57:01Z
--- tags: - text-to-image - flux - lora - diffusers - template:sd-lora - fluxgym base_model: black-forest-labs/FLUX.1-dev instance_prompt: girllt license: other license_name: flux-1-dev-non-commercial-license license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md --- # girllt A Flux LoRA trained on a local computer with [Fluxgym](https://github.com/cocktailpeanut/fluxgym) <Gallery /> ## Trigger words You should use `girllt` to trigger the image generation. ## Download model and use it with ComfyUI, AUTOMATIC1111, SD.Next, Invoke AI, Forge, etc. Weights for this model are available in Safetensors format.
mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF
mradermacher
2024-10-13T10:29:05Z
65
0
transformers
[ "transformers", "gguf", "text-generation-inference", "unsloth", "llama", "trl", "sft", "en", "fi", "dataset:wikimedia/wikipedia", "base_model:mpasila/Llama-3.2-Finnish-Wikipedia-1B", "base_model:quantized:mpasila/Llama-3.2-Finnish-Wikipedia-1B", "license:llama3.2", "endpoints_compatible", "region:us" ]
null
2024-10-13T10:25:47Z
--- base_model: mpasila/Llama-3.2-Finnish-Wikipedia-1B datasets: - wikimedia/wikipedia language: - en - fi library_name: transformers license: llama3.2 quantized_by: mradermacher tags: - text-generation-inference - transformers - unsloth - llama - trl - sft --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/mpasila/Llama-3.2-Finnish-Wikipedia-1B <!-- provided-files --> weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q2_K.gguf) | Q2_K | 0.8 | | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q3_K_S.gguf) | Q3_K_S | 0.9 | | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q3_K_M.gguf) | Q3_K_M | 0.9 | lower quality | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q3_K_L.gguf) | Q3_K_L | 0.9 | | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.IQ4_XS.gguf) | IQ4_XS | 1.0 | | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q4_K_S.gguf) | Q4_K_S | 1.0 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q4_K_M.gguf) | Q4_K_M | 1.1 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q5_K_S.gguf) | Q5_K_S | 1.2 | | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q5_K_M.gguf) | Q5_K_M | 1.2 | | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q6_K.gguf) | Q6_K | 1.3 | very good quality | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.Q8_0.gguf) | Q8_0 | 1.7 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-Finnish-Wikipedia-1B-GGUF/resolve/main/Llama-3.2-Finnish-Wikipedia-1B.f16.gguf) | f16 | 3.1 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF
mradermacher
2024-10-13T10:28:51Z
15
0
transformers
[ "transformers", "gguf", "en", "dataset:mosaicml/instruct-v3", "base_model:Youlln/ECE-PRYMMAL1B-FT-V1", "base_model:quantized:Youlln/ECE-PRYMMAL1B-FT-V1", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T10:23:31Z
--- base_model: Youlln/ECE-PRYMMAL1B-FT-V1 datasets: - mosaicml/instruct-v3 language: - en library_name: transformers license: apache-2.0 quantized_by: mradermacher --- ## About <!-- ### quantize_version: 2 --> <!-- ### output_tensor_quantised: 1 --> <!-- ### convert_type: hf --> <!-- ### vocab_type: --> <!-- ### tags: --> static quants of https://huggingface.co/Youlln/ECE-PRYMMAL1B-FT-V1 <!-- provided-files --> weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion. ## Usage If you are unsure how to use GGUF files, refer to one of [TheBloke's READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for more details, including on how to concatenate multi-part files. ## Provided Quants (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) | Link | Type | Size/GB | Notes | |:-----|:-----|--------:|:------| | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q2_K.gguf) | Q2_K | 0.8 | | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q3_K_S.gguf) | Q3_K_S | 0.9 | | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q3_K_M.gguf) | Q3_K_M | 0.9 | lower quality | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q3_K_L.gguf) | Q3_K_L | 1.0 | | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.IQ4_XS.gguf) | IQ4_XS | 1.0 | | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q4_K_S.gguf) | Q4_K_S | 1.0 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q4_K_M.gguf) | Q4_K_M | 1.1 | fast, recommended | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q5_K_S.gguf) | Q5_K_S | 1.2 | | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q5_K_M.gguf) | Q5_K_M | 1.2 | | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q6_K.gguf) | Q6_K | 1.4 | very good quality | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.Q8_0.gguf) | Q8_0 | 1.7 | fast, best quality | | [GGUF](https://huggingface.co/mradermacher/ECE-PRYMMAL1B-FT-V1-GGUF/resolve/main/ECE-PRYMMAL1B-FT-V1.f16.gguf) | f16 | 3.2 | 16 bpw, overkill | Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better): ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png) And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9 ## FAQ / Model Request See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized. ## Thanks I thank my company, [nethype GmbH](https://www.nethype.de/), for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. <!-- end -->
QuantFactory/mamba-2.8b-hf-GGUF
QuantFactory
2024-10-13T10:24:17Z
163
2
transformers
[ "transformers", "gguf", "endpoints_compatible", "region:us" ]
null
2024-10-13T10:11:22Z
--- library_name: transformers tags: [] --- [![QuantFactory Banner](https://lh7-rt.googleusercontent.com/docsz/AD_4nXeiuCm7c8lEwEJuRey9kiVZsRn2W-b4pWlu3-X534V3YmVuVc2ZL-NXg2RkzSOOS2JXGHutDuyyNAUtdJI65jGTo8jT9Y99tMi4H4MqL44Uc5QKG77B0d6-JfIkZHFaUA71-RtjyYZWVIhqsNZcx8-OMaA?key=xt3VSDoCbmTY7o-cwwOFwQ)](https://hf.co/QuantFactory) # QuantFactory/mamba-2.8b-hf-GGUF This is quantized version of [state-spaces/mamba-2.8b-hf](https://huggingface.co/state-spaces/mamba-2.8b-hf) created using llama.cpp # Original Model Card # Mamba <!-- Provide a quick summary of what the model is/does. --> This repository contains the `transfromers` compatible `mamba-2.8b`. The checkpoints are untouched, but the full `config.json` and tokenizer are pushed to this repo. # Usage You need to install `transformers` from `main` until `transformers=4.39.0` is released. ```bash pip install git+https://github.com/huggingface/transformers@main ``` We also recommend you to install both `causal_conv_1d` and `mamba-ssm` using: ```bash pip install causal-conv1d>=1.2.0 pip install mamba-ssm ``` If any of these two is not installed, the "eager" implementation will be used. Otherwise the more optimised `cuda` kernels will be used. ## Generation You can use the classic `generate` API: ```python >>> from transformers import MambaConfig, MambaForCausalLM, AutoTokenizer >>> import torch >>> tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-2.8b-hf") >>> model = MambaForCausalLM.from_pretrained("state-spaces/mamba-2.8b-hf") >>> input_ids = tokenizer("Hey how are you doing?", return_tensors="pt")["input_ids"] >>> out = model.generate(input_ids, max_new_tokens=10) >>> print(tokenizer.batch_decode(out)) ["Hey how are you doing?\n\nI'm doing great.\n\nI"] ``` ## PEFT finetuning example In order to finetune using the `peft` library, we recommend keeping the model in float32! ```python from datasets import load_dataset from trl import SFTTrainer from peft import LoraConfig from transformers import AutoTokenizer, AutoModelForCausalLM, TrainingArguments tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-2.8b-hf") model = AutoModelForCausalLM.from_pretrained("state-spaces/mamba-2.8b-hf") dataset = load_dataset("Abirate/english_quotes", split="train") training_args = TrainingArguments( output_dir="./results", num_train_epochs=3, per_device_train_batch_size=4, logging_dir='./logs', logging_steps=10, learning_rate=2e-3 ) lora_config = LoraConfig( r=8, target_modules=["x_proj", "embeddings", "in_proj", "out_proj"], task_type="CAUSAL_LM", bias="none" ) trainer = SFTTrainer( model=model, tokenizer=tokenizer, args=training_args, peft_config=lora_config, train_dataset=dataset, dataset_text_field="quote", ) trainer.train() ```
Gummybear05/whisper-small-E10_speed_pause
Gummybear05
2024-10-13T10:17:45Z
7
0
transformers
[ "transformers", "tensorboard", "safetensors", "whisper", "automatic-speech-recognition", "hf-asr-leaderboard", "generated_from_trainer", "hi", "dataset:aihub_adult_baseline", "base_model:openai/whisper-small", "base_model:finetune:openai/whisper-small", "license:apache-2.0", "endpoints_compatible", "region:us" ]
automatic-speech-recognition
2024-10-13T08:34:40Z
--- library_name: transformers language: - hi license: apache-2.0 base_model: openai/whisper-small tags: - hf-asr-leaderboard - generated_from_trainer datasets: - aihub_adult_baseline model-index: - name: whisper-small-Yfreq_speed_pause results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # whisper-small-Yfreq_speed_pause This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the aihub old10 adult speed pause changed dataset. It achieves the following results on the evaluation set: - Loss: 0.2307 - Cer: 7.4836 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 16 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 50 - num_epochs: 2 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Cer | |:-------------:|:------:|:----:|:---------------:|:------:| | 0.4388 | 0.1289 | 100 | 0.3025 | 7.1957 | | 0.2656 | 0.2579 | 200 | 0.2567 | 6.9725 | | 0.2193 | 0.3868 | 300 | 0.2541 | 7.3073 | | 0.2021 | 0.5158 | 400 | 0.2426 | 7.2075 | | 0.1913 | 0.6447 | 500 | 0.2465 | 7.6539 | | 0.1617 | 0.7737 | 600 | 0.2449 | 7.0371 | | 0.1561 | 0.9026 | 700 | 0.2403 | 7.1546 | | 0.0734 | 1.0316 | 800 | 0.2329 | 7.5070 | | 0.0571 | 1.1605 | 900 | 0.2329 | 7.4424 | | 0.0631 | 1.2895 | 1000 | 0.2330 | 7.5952 | | 0.06 | 1.4184 | 1100 | 0.2358 | 7.5129 | | 0.0655 | 1.5474 | 1200 | 0.2305 | 7.4659 | | 0.0611 | 1.6763 | 1300 | 0.2303 | 7.3308 | | 0.0574 | 1.8053 | 1400 | 0.2308 | 7.5482 | | 0.0615 | 1.9342 | 1500 | 0.2307 | 7.4836 | ### Framework versions - Transformers 4.46.0.dev0 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.20.1
pandafm/donut-es
pandafm
2024-10-13T10:10:40Z
19
0
transformers
[ "transformers", "safetensors", "vision-encoder-decoder", "image-text-to-text", "donut", "es", "base_model:naver-clova-ix/donut-base-finetuned-cord-v2", "base_model:finetune:naver-clova-ix/donut-base-finetuned-cord-v2", "license:mit", "endpoints_compatible", "region:us" ]
image-text-to-text
2024-05-07T13:52:42Z
--- library_name: transformers tags: - donut license: mit language: - es base_model: - naver-clova-ix/donut-base-finetuned-cord-v2 --- # Model Card: Donut Model for Ticket Parsing ## Model Description This is a fine-tuned version of the [Donut](https://huggingface.co/naver-clova-ix/donut-base-finetuned-cord-v2) architecture, specifically tailored for parsing retail receipts. Donut is a transformer-based model designed for document understanding, and it performs OCR-free parsing by directly processing images into structured JSON outputs. This implementation was fine-tuned using a custom dataset of artificial and real receipts. ### Use Case This model is intended to be used for parsing receipts into structured data, extracting information such as item names, quantities, prices, taxes, and total amounts directly from image inputs. ## Dataset The model was trained on a mixture of synthetic and real-world receipts: - **Artificial Receipts**: Generated using a custom tool inspired by SynthDoG and built with OpenCV. The tool simulates various real-world conditions (e.g., Gaussian noise, wrinkles, luminance variations) to enhance the robustness of the model. - **Real Receipts**: A manually parsed dataset of 704 receipts, including a validation set of 200 receipts. ### Data Creation Process The artificial receipts were generated using a combination of background images, fonts, and custom templates to mimic real-world conditions, ensuring the model can handle various types of distortions such as noise, wrinkles, and lighting changes. The real receipts were annotated manually using a custom tool based on the Marimo app, which allowed for structured annotation of receipt elements. ## Training Details - **Hardware**: The model was trained using Google Colab Pro. - **Training Steps**: The model was trained in three main steps of 10 epochs each, totaling 30 epochs. - **Loss Function**: The model was trained using a combination of Levenshtein edit-distance for string similarity and nTED (normalized Tree Edit Distance) for accuracy in tree-based data structures. - **Performance**: The model showed significant improvements when trained with a mix of artificial and real receipts, achieving a validation accuracy of 0.98 and a test accuracy of 0.70. ### Results The model was tested on both validation and test datasets, achieving the following results: - **Validation Accuracy**: 98.37% (final fine-tuned model) - **Test Accuracy**: 69.63% (final fine-tuned model) ## Limitations - **Synthetic Data**: Although artificial receipts helped improve performance, the model may still struggle with unseen or very complex receipt formats that weren't part of the training dataset. - **Real-world Deployment**: Further fine-tuning might be necessary to adapt the model to new types of receipts or different languages. ## Ethical Considerations - **Privacy**: Care should be taken when using this model on personal or sensitive financial data. Ensure compliance with local privacy laws and regulations. - **Bias**: The model was trained on a limited set of receipts, which could result in biases toward certain types of stores or receipt formats. ## How to Use This model is available on Hugging Face and can be used as follows: ```python from transformers import DonutProcessor, VisionEncoderDecoderModel from PIL import Image import json import torch import re # Load model and processor print("Loading Donut model...") processor = DonutProcessor.from_pretrained("pandafm/donut-es") model = VisionEncoderDecoderModel.from_pretrained("pandafm/donut-es") if torch.cuda.is_available(): device = torch.device("cuda") model.to(device) else: model.encoder.to(torch.bfloat16) print("Donut model loaded.") # Open image of a receipt image = Image.open("path_to_receipt_image.jpg") # Process image and generate JSON output pixel_values = processor(image, return_tensors="pt").pixel_values if torch.cuda.is_available(): pixel_values = pixel_values.to(device) else: pixel_values = pixel_values.to(torch.bfloat16) # Convert output to JSON task_prompt = "<s_cord-v2>" decoder_input_ids = processor.tokenizer(task_prompt, add_special_tokens=False, return_tensors="pt").input_ids decoder_input_ids = decoder_input_ids.to(device) # autoregressively generate sequence result = model.generate( pixel_values, decoder_input_ids=decoder_input_ids, max_length=model.decoder.config.max_position_embeddings, pad_token_id=processor.tokenizer.pad_token_id, eos_token_id=processor.tokenizer.eos_token_id, bad_words_ids=[[processor.tokenizer.unk_token_id]], return_dict_in_generate=True, ) seq = processor.batch_decode(result.sequences)[0] seq = seq.replace(processor.tokenizer.eos_token, "").replace(processor.tokenizer.pad_token, "") seq = re.sub(r"<.*?>", "", seq, count=1).strip() # remove first task start token seq = processor.token2json(seq) ``` ## Acknowledgements This model was fine-tuned as part of a research project for a Bachelor's Degree, leveraging the Donut architecture and integrating tools like OpenCV for data generation. The final dataset included both synthetic and real-world receipts to improve robustness in parsing. ## Citation @thesis{pandafm2024DonutES, author = {David Florez Mazuera}, title = {Ticket Parser}, school = {Universidad de Murcia}, year = {2024}, address = {Murcia, España}, month = {June}, type = {Bachelor's thesis}, note = {Gines García Mateos}, url = {}, keywords = {donut, transformers, fine-tune}, }
QuantFactory/qwen2.5-7b-ins-v3-GGUF
QuantFactory
2024-10-13T10:02:15Z
21
2
null
[ "gguf", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T09:25:20Z
--- license: apache-2.0 --- [![QuantFactory Banner](https://lh7-rt.googleusercontent.com/docsz/AD_4nXeiuCm7c8lEwEJuRey9kiVZsRn2W-b4pWlu3-X534V3YmVuVc2ZL-NXg2RkzSOOS2JXGHutDuyyNAUtdJI65jGTo8jT9Y99tMi4H4MqL44Uc5QKG77B0d6-JfIkZHFaUA71-RtjyYZWVIhqsNZcx8-OMaA?key=xt3VSDoCbmTY7o-cwwOFwQ)](https://hf.co/QuantFactory) # QuantFactory/qwen2.5-7b-ins-v3-GGUF This is quantized version of [happzy2633/qwen2.5-7b-ins-v3](https://huggingface.co/happzy2633/qwen2.5-7b-ins-v3) created using llama.cpp # Original Model Card
cuongdev/pxhienneew2
cuongdev
2024-10-13T09:59:06Z
30
0
diffusers
[ "diffusers", "text-to-image", "stable-diffusion", "license:creativeml-openrail-m", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionPipeline", "region:us" ]
text-to-image
2024-10-13T09:55:40Z
--- license: creativeml-openrail-m tags: - text-to-image - stable-diffusion --- ### pxhienneew2 Dreambooth model trained by cuongdev with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) Sample pictures of this concept:
Ya211/llamaEXL
Ya211
2024-10-13T09:53:44Z
6
0
null
[ "safetensors", "llama", "facebook", "meta", "pytorch", "llama-3", "text-generation", "conversational", "en", "de", "fr", "it", "pt", "hi", "es", "th", "arxiv:2204.05149", "base_model:meta-llama/Llama-3.1-8B", "base_model:quantized:meta-llama/Llama-3.1-8B", "license:llama3.1", "5-bit", "exl2", "region:us" ]
text-generation
2024-10-13T09:47:48Z
--- language: - en - de - fr - it - pt - hi - es - th license: llama3.1 base_model: meta-llama/Meta-Llama-3.1-8B pipeline_tag: text-generation tags: - facebook - meta - pytorch - llama - llama-3 extra_gated_prompt: "### LLAMA 3.1 COMMUNITY LICENSE AGREEMENT\nLlama 3.1 Version\ \ Release Date: July 23, 2024\n\"Agreement\" means the terms and conditions for\ \ use, reproduction, distribution and modification of the Llama Materials set forth\ \ herein.\n\"Documentation\" means the specifications, manuals and documentation\ \ accompanying Llama 3.1 distributed by Meta at https://llama.meta.com/doc/overview.\n\ \"Licensee\" or \"you\" means you, or your employer or any other person or entity\ \ (if you are entering into this Agreement on such person or entity’s behalf), of\ \ the age required under applicable laws, rules or regulations to provide legal\ \ consent and that has legal authority to bind your employer or such other person\ \ or entity if you are entering in this Agreement on their behalf.\n\"Llama 3.1\"\ \ means the foundational large language models and software and algorithms, including\ \ machine-learning model code, trained model weights, inference-enabling code, training-enabling\ \ code, fine-tuning enabling code and other elements of the foregoing distributed\ \ by Meta at https://llama.meta.com/llama-downloads.\n\"Llama Materials\" means,\ \ collectively, Meta’s proprietary Llama 3.1 and Documentation (and any portion\ \ thereof) made available under this Agreement.\n\"Meta\" or \"we\" means Meta Platforms\ \ Ireland Limited (if you are located in or, if you are an entity, your principal\ \ place of business is in the EEA or Switzerland) and Meta Platforms, Inc. (if you\ \ are located outside of the EEA or Switzerland).\n \n1. License Rights and Redistribution.\n\ a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable\ \ and royalty-free limited license under Meta’s intellectual property or other rights\ \ owned by Meta embodied in the Llama Materials to use, reproduce, distribute, copy,\ \ create derivative works of, and make modifications to the Llama Materials.\nb.\ \ Redistribution and Use.\ni. If you distribute or make available the Llama Materials\ \ (or any derivative works thereof), or a product or service (including another\ \ AI model) that contains any of them, you shall (A) provide a copy of this Agreement\ \ with any such Llama Materials; and (B) prominently display “Built with Llama”\ \ on a related website, user interface, blogpost, about page, or product documentation.\ \ If you use the Llama Materials or any outputs or results of the Llama Materials\ \ to create, train, fine tune, or otherwise improve an AI model, which is distributed\ \ or made available, you shall also include “Llama” at the beginning of any such\ \ AI model name.\nii. If you receive Llama Materials, or any derivative works thereof,\ \ from a Licensee as part of an integrated end user product, then Section 2 of\ \ this Agreement will not apply to you.\niii. You must retain in all copies of the\ \ Llama Materials that you distribute the following attribution notice within a\ \ “Notice” text file distributed as a part of such copies: “Llama 3.1 is licensed\ \ under the Llama 3.1 Community License, Copyright © Meta Platforms, Inc. All Rights\ \ Reserved.”\niv. Your use of the Llama Materials must comply with applicable laws\ \ and regulations (including trade compliance laws and regulations) and adhere to\ \ the Acceptable Use Policy for the Llama Materials (available at https://llama.meta.com/llama3_1/use-policy),\ \ which is hereby incorporated by reference into this Agreement.\n2. Additional\ \ Commercial Terms. If, on the Llama 3.1 version release date, the monthly active\ \ users of the products or services made available by or for Licensee, or Licensee’s\ \ affiliates, is greater than 700 million monthly active users in the preceding\ \ calendar month, you must request a license from Meta, which Meta may grant to\ \ you in its sole discretion, and you are not authorized to exercise any of the\ \ rights under this Agreement unless or until Meta otherwise expressly grants you\ \ such rights.\n3. Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE\ \ LLAMA MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS”\ \ BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF ANY\ \ KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES\ \ OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.\ \ YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING\ \ THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED WITH YOUR USE OF THE LLAMA\ \ MATERIALS AND ANY OUTPUT AND RESULTS.\n4. Limitation of Liability. IN NO EVENT\ \ WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY, WHETHER IN\ \ CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS\ \ AGREEMENT, FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL,\ \ EXEMPLARY OR PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED\ \ OF THE POSSIBILITY OF ANY OF THE FOREGOING.\n5. Intellectual Property.\na. No\ \ trademark licenses are granted under this Agreement, and in connection with the\ \ Llama Materials, neither Meta nor Licensee may use any name or mark owned by or\ \ associated with the other or any of its affiliates, except as required for reasonable\ \ and customary use in describing and redistributing the Llama Materials or as set\ \ forth in this Section 5(a). Meta hereby grants you a license to use “Llama” (the\ \ “Mark”) solely as required to comply with the last sentence of Section 1.b.i.\ \ You will comply with Meta’s brand guidelines (currently accessible at https://about.meta.com/brand/resources/meta/company-brand/\ \ ). All goodwill arising out of your use of the Mark will inure to the benefit\ \ of Meta.\nb. Subject to Meta’s ownership of Llama Materials and derivatives made\ \ by or for Meta, with respect to any derivative works and modifications of the\ \ Llama Materials that are made by you, as between you and Meta, you are and will\ \ be the owner of such derivative works and modifications.\nc. If you institute\ \ litigation or other proceedings against Meta or any entity (including a cross-claim\ \ or counterclaim in a lawsuit) alleging that the Llama Materials or Llama 3.1 outputs\ \ or results, or any portion of any of the foregoing, constitutes infringement of\ \ intellectual property or other rights owned or licensable by you, then any licenses\ \ granted to you under this Agreement shall terminate as of the date such litigation\ \ or claim is filed or instituted. You will indemnify and hold harmless Meta from\ \ and against any claim by any third party arising out of or related to your use\ \ or distribution of the Llama Materials.\n6. Term and Termination. The term of\ \ this Agreement will commence upon your acceptance of this Agreement or access\ \ to the Llama Materials and will continue in full force and effect until terminated\ \ in accordance with the terms and conditions herein. Meta may terminate this Agreement\ \ if you are in breach of any term or condition of this Agreement. Upon termination\ \ of this Agreement, you shall delete and cease use of the Llama Materials. Sections\ \ 3, 4 and 7 shall survive the termination of this Agreement.\n7. Governing Law\ \ and Jurisdiction. This Agreement will be governed and construed under the laws\ \ of the State of California without regard to choice of law principles, and the\ \ UN Convention on Contracts for the International Sale of Goods does not apply\ \ to this Agreement. The courts of California shall have exclusive jurisdiction\ \ of any dispute arising out of this Agreement.\n### Llama 3.1 Acceptable Use Policy\n\ Meta is committed to promoting safe and fair use of its tools and features, including\ \ Llama 3.1. If you access or use Llama 3.1, you agree to this Acceptable Use Policy\ \ (“Policy”). The most recent copy of this policy can be found at [https://llama.meta.com/llama3_1/use-policy](https://llama.meta.com/llama3_1/use-policy)\n\ #### Prohibited Uses\nWe want everyone to use Llama 3.1 safely and responsibly.\ \ You agree you will not use, or allow others to use, Llama 3.1 to:\n 1. Violate\ \ the law or others’ rights, including to:\n 1. Engage in, promote, generate,\ \ contribute to, encourage, plan, incite, or further illegal or unlawful activity\ \ or content, such as:\n 1. Violence or terrorism\n 2. Exploitation\ \ or harm to children, including the solicitation, creation, acquisition, or dissemination\ \ of child exploitative content or failure to report Child Sexual Abuse Material\n\ \ 3. Human trafficking, exploitation, and sexual violence\n 4. The\ \ illegal distribution of information or materials to minors, including obscene\ \ materials, or failure to employ legally required age-gating in connection with\ \ such information or materials.\n 5. Sexual solicitation\n 6. Any\ \ other criminal activity\n 3. Engage in, promote, incite, or facilitate the\ \ harassment, abuse, threatening, or bullying of individuals or groups of individuals\n\ \ 4. Engage in, promote, incite, or facilitate discrimination or other unlawful\ \ or harmful conduct in the provision of employment, employment benefits, credit,\ \ housing, other economic benefits, or other essential goods and services\n 5.\ \ Engage in the unauthorized or unlicensed practice of any profession including,\ \ but not limited to, financial, legal, medical/health, or related professional\ \ practices\n 6. Collect, process, disclose, generate, or infer health, demographic,\ \ or other sensitive personal or private information about individuals without rights\ \ and consents required by applicable laws\n 7. Engage in or facilitate any action\ \ or generate any content that infringes, misappropriates, or otherwise violates\ \ any third-party rights, including the outputs or results of any products or services\ \ using the Llama Materials\n 8. Create, generate, or facilitate the creation\ \ of malicious code, malware, computer viruses or do anything else that could disable,\ \ overburden, interfere with or impair the proper working, integrity, operation\ \ or appearance of a website or computer system\n2. Engage in, promote, incite,\ \ facilitate, or assist in the planning or development of activities that present\ \ a risk of death or bodily harm to individuals, including use of Llama 3.1 related\ \ to the following:\n 1. Military, warfare, nuclear industries or applications,\ \ espionage, use for materials or activities that are subject to the International\ \ Traffic Arms Regulations (ITAR) maintained by the United States Department of\ \ State\n 2. Guns and illegal weapons (including weapon development)\n 3.\ \ Illegal drugs and regulated/controlled substances\n 4. Operation of critical\ \ infrastructure, transportation technologies, or heavy machinery\n 5. Self-harm\ \ or harm to others, including suicide, cutting, and eating disorders\n 6. Any\ \ content intended to incite or promote violence, abuse, or any infliction of bodily\ \ harm to an individual\n3. Intentionally deceive or mislead others, including use\ \ of Llama 3.1 related to the following:\n 1. Generating, promoting, or furthering\ \ fraud or the creation or promotion of disinformation\n 2. Generating, promoting,\ \ or furthering defamatory content, including the creation of defamatory statements,\ \ images, or other content\n 3. Generating, promoting, or further distributing\ \ spam\n 4. Impersonating another individual without consent, authorization,\ \ or legal right\n 5. Representing that the use of Llama 3.1 or outputs are human-generated\n\ \ 6. Generating or facilitating false online engagement, including fake reviews\ \ and other means of fake online engagement\n4. Fail to appropriately disclose to\ \ end users any known dangers of your AI system\nPlease report any violation of\ \ this Policy, software “bug,” or other problems that could lead to a violation\ \ of this Policy through one of the following means:\n * Reporting issues with\ \ the model: [https://github.com/meta-llama/llama-models/issues](https://github.com/meta-llama/llama-models/issues)\n\ \ * Reporting risky content generated by the model:\n developers.facebook.com/llama_output_feedback\n\ \ * Reporting bugs and security concerns: facebook.com/whitehat/info\n * Reporting\ \ violations of the Acceptable Use Policy or unlicensed uses of Meta Llama 3: [email protected]" extra_gated_fields: First Name: text Last Name: text Date of birth: date_picker Country: country Affiliation: text Job title: type: select options: - Student - Research Graduate - AI researcher - AI developer/engineer - Reporter - Other geo: ip_location ? By clicking Submit below I accept the terms of the license and acknowledge that the information I provide will be collected stored processed and shared in accordance with the Meta Privacy Policy : checkbox extra_gated_description: The information you provide will be collected, stored, processed and shared in accordance with the [Meta Privacy Policy](https://www.facebook.com/privacy/policy/). extra_gated_button_content: Submit --- ## Model Information The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks. **Model developer**: Meta **Model Architecture:** Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety. <table> <tr> <td> </td> <td><strong>Training Data</strong> </td> <td><strong>Params</strong> </td> <td><strong>Input modalities</strong> </td> <td><strong>Output modalities</strong> </td> <td><strong>Context length</strong> </td> <td><strong>GQA</strong> </td> <td><strong>Token count</strong> </td> <td><strong>Knowledge cutoff</strong> </td> </tr> <tr> <td rowspan="3" >Llama 3.1 (text only) </td> <td rowspan="3" >A new mix of publicly available online data. </td> <td>8B </td> <td>Multilingual Text </td> <td>Multilingual Text and code </td> <td>128k </td> <td>Yes </td> <td rowspan="3" >15T+ </td> <td rowspan="3" >December 2023 </td> </tr> <tr> <td>70B </td> <td>Multilingual Text </td> <td>Multilingual Text and code </td> <td>128k </td> <td>Yes </td> </tr> <tr> <td>405B </td> <td>Multilingual Text </td> <td>Multilingual Text and code </td> <td>128k </td> <td>Yes </td> </tr> </table> **Supported languages:** English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai. **Llama 3.1 family of models**. Token counts refer to pretraining data only. All model versions use Grouped-Query Attention (GQA) for improved inference scalability. **Model Release Date:** July 23, 2024. **Status:** This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback. **License:** A custom commercial license, the Llama 3.1 Community License, is available at: [https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/LICENSE](https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/LICENSE) Where to send questions or comments about the model Instructions on how to provide feedback or comments on the model can be found in the model [README](https://github.com/meta-llama/llama3). For more technical information about generation parameters and recipes for how to use Llama 3.1 in applications, please go [here](https://github.com/meta-llama/llama-recipes). ## Intended Use **Intended Use Cases** Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 model collection also supports the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. The Llama 3.1 Community License allows for these use cases. **Out-of-scope** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3.1 Community License. Use in languages beyond those explicitly referenced as supported in this model card**. **<span style="text-decoration:underline;">Note</span>: Llama 3.1 has been trained on a broader collection of languages than the 8 supported languages. Developers may fine-tune Llama 3.1 models for languages beyond the 8 supported languages provided they comply with the Llama 3.1 Community License and the Acceptable Use Policy and in such cases are responsible for ensuring that any uses of Llama 3.1 in additional languages is done in a safe and responsible manner. ## How to use This repository contains two versions of Meta-Llama-3.1-8B-Instruct, for use with transformers and with the original `llama` codebase. ### Use with transformers Starting with `transformers >= 4.43.0` onward, you can run conversational inference using the Transformers `pipeline` abstraction or by leveraging the Auto classes with the `generate()` function. Make sure to update your transformers installation via `pip install --upgrade transformers`. ```python import transformers import torch model_id = "meta-llama/Meta-Llama-3.1-8B-Instruct" pipeline = transformers.pipeline( "text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device_map="auto", ) messages = [ {"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"}, {"role": "user", "content": "Who are you?"}, ] outputs = pipeline( messages, max_new_tokens=256, ) print(outputs[0]["generated_text"][-1]) ``` Note: You can also find detailed recipes on how to use the model locally, with `torch.compile()`, assisted generations, quantised and more at [`huggingface-llama-recipes`](https://github.com/huggingface/huggingface-llama-recipes) ### Tool use with transformers LLaMA-3.1 supports multiple tool use formats. You can see a full guide to prompt formatting [here](https://llama.meta.com/docs/model-cards-and-prompt-formats/llama3_1/). Tool use is also supported through [chat templates](https://huggingface.co/docs/transformers/main/chat_templating#advanced-tool-use--function-calling) in Transformers. Here is a quick example showing a single simple tool: ```python # First, define a tool def get_current_temperature(location: str) -> float: """ Get the current temperature at a location. Args: location: The location to get the temperature for, in the format "City, Country" Returns: The current temperature at the specified location in the specified units, as a float. """ return 22. # A real function should probably actually get the temperature! # Next, create a chat and apply the chat template messages = [ {"role": "system", "content": "You are a bot that responds to weather queries."}, {"role": "user", "content": "Hey, what's the temperature in Paris right now?"} ] inputs = tokenizer.apply_chat_template(messages, tools=[get_current_temperature], add_generation_prompt=True) ``` You can then generate text from this input as normal. If the model generates a tool call, you should add it to the chat like so: ```python tool_call = {"name": "get_current_temperature", "arguments": {"location": "Paris, France"}} messages.append({"role": "assistant", "tool_calls": [{"type": "function", "function": tool_call}]}) ``` and then call the tool and append the result, with the `tool` role, like so: ```python messages.append({"role": "tool", "name": "get_current_temperature", "content": "22.0"}) ``` After that, you can `generate()` again to let the model use the tool result in the chat. Note that this was a very brief introduction to tool calling - for more information, see the [LLaMA prompt format docs](https://llama.meta.com/docs/model-cards-and-prompt-formats/llama3_1/) and the Transformers [tool use documentation](https://huggingface.co/docs/transformers/main/chat_templating#advanced-tool-use--function-calling). ### Use with `llama` Please, follow the instructions in the [repository](https://github.com/meta-llama/llama) To download Original checkpoints, see the example command below leveraging `huggingface-cli`: ``` huggingface-cli download meta-llama/Meta-Llama-3.1-8B-Instruct --include "original/*" --local-dir Meta-Llama-3.1-8B-Instruct ``` ## Hardware and Software **Training Factors** We used custom training libraries, Meta's custom built GPU cluster, and production infrastructure for pretraining. Fine-tuning, annotation, and evaluation were also performed on production infrastructure. **Training utilized a cumulative of** 39.3M GPU hours of computation on H100-80GB (TDP of 700W) type hardware, per the table below. Training time is the total GPU time required for training each model and power consumption is the peak power capacity per GPU device used, adjusted for power usage efficiency. **Training Greenhouse Gas Emissions** Estimated total location-based greenhouse gas emissions were **11,390** tons CO2eq for training. Since 2020, Meta has maintained net zero greenhouse gas emissions in its global operations and matched 100% of its electricity use with renewable energy, therefore the total market-based greenhouse gas emissions for training were 0 tons CO2eq. <table> <tr> <td> </td> <td><strong>Training Time (GPU hours)</strong> </td> <td><strong>Training Power Consumption (W)</strong> </td> <td><strong>Training Location-Based Greenhouse Gas Emissions</strong> <p> <strong>(tons CO2eq)</strong> </td> <td><strong>Training Market-Based Greenhouse Gas Emissions</strong> <p> <strong>(tons CO2eq)</strong> </td> </tr> <tr> <td>Llama 3.1 8B </td> <td>1.46M </td> <td>700 </td> <td>420 </td> <td>0 </td> </tr> <tr> <td>Llama 3.1 70B </td> <td>7.0M </td> <td>700 </td> <td>2,040 </td> <td>0 </td> </tr> <tr> <td>Llama 3.1 405B </td> <td>30.84M </td> <td>700 </td> <td>8,930 </td> <td>0 </td> </tr> <tr> <td>Total </td> <td>39.3M <td> <ul> </ul> </td> <td>11,390 </td> <td>0 </td> </tr> </table> The methodology used to determine training energy use and greenhouse gas emissions can be found [here](https://arxiv.org/pdf/2204.05149). Since Meta is openly releasing these models, the training energy use and greenhouse gas emissions will not be incurred by others. ## Training Data **Overview:** Llama 3.1 was pretrained on ~15 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over 25M synthetically generated examples. **Data Freshness:** The pretraining data has a cutoff of December 2023. ## Benchmark scores In this section, we report the results for Llama 3.1 models on standard automatic benchmarks. For all the evaluations, we use our internal evaluations library. ### Base pretrained models <table> <tr> <td><strong>Category</strong> </td> <td><strong>Benchmark</strong> </td> <td><strong># Shots</strong> </td> <td><strong>Metric</strong> </td> <td><strong>Llama 3 8B</strong> </td> <td><strong>Llama 3.1 8B</strong> </td> <td><strong>Llama 3 70B</strong> </td> <td><strong>Llama 3.1 70B</strong> </td> <td><strong>Llama 3.1 405B</strong> </td> </tr> <tr> <td rowspan="7" >General </td> <td>MMLU </td> <td>5 </td> <td>macro_avg/acc_char </td> <td>66.7 </td> <td>66.7 </td> <td>79.5 </td> <td>79.3 </td> <td>85.2 </td> </tr> <tr> <td>MMLU-Pro (CoT) </td> <td>5 </td> <td>macro_avg/acc_char </td> <td>36.2 </td> <td>37.1 </td> <td>55.0 </td> <td>53.8 </td> <td>61.6 </td> </tr> <tr> <td>AGIEval English </td> <td>3-5 </td> <td>average/acc_char </td> <td>47.1 </td> <td>47.8 </td> <td>63.0 </td> <td>64.6 </td> <td>71.6 </td> </tr> <tr> <td>CommonSenseQA </td> <td>7 </td> <td>acc_char </td> <td>72.6 </td> <td>75.0 </td> <td>83.8 </td> <td>84.1 </td> <td>85.8 </td> </tr> <tr> <td>Winogrande </td> <td>5 </td> <td>acc_char </td> <td>- </td> <td>60.5 </td> <td>- </td> <td>83.3 </td> <td>86.7 </td> </tr> <tr> <td>BIG-Bench Hard (CoT) </td> <td>3 </td> <td>average/em </td> <td>61.1 </td> <td>64.2 </td> <td>81.3 </td> <td>81.6 </td> <td>85.9 </td> </tr> <tr> <td>ARC-Challenge </td> <td>25 </td> <td>acc_char </td> <td>79.4 </td> <td>79.7 </td> <td>93.1 </td> <td>92.9 </td> <td>96.1 </td> </tr> <tr> <td>Knowledge reasoning </td> <td>TriviaQA-Wiki </td> <td>5 </td> <td>em </td> <td>78.5 </td> <td>77.6 </td> <td>89.7 </td> <td>89.8 </td> <td>91.8 </td> </tr> <tr> <td rowspan="4" >Reading comprehension </td> <td>SQuAD </td> <td>1 </td> <td>em </td> <td>76.4 </td> <td>77.0 </td> <td>85.6 </td> <td>81.8 </td> <td>89.3 </td> </tr> <tr> <td>QuAC (F1) </td> <td>1 </td> <td>f1 </td> <td>44.4 </td> <td>44.9 </td> <td>51.1 </td> <td>51.1 </td> <td>53.6 </td> </tr> <tr> <td>BoolQ </td> <td>0 </td> <td>acc_char </td> <td>75.7 </td> <td>75.0 </td> <td>79.0 </td> <td>79.4 </td> <td>80.0 </td> </tr> <tr> <td>DROP (F1) </td> <td>3 </td> <td>f1 </td> <td>58.4 </td> <td>59.5 </td> <td>79.7 </td> <td>79.6 </td> <td>84.8 </td> </tr> </table> ### Instruction tuned models <table> <tr> <td><strong>Category</strong> </td> <td><strong>Benchmark</strong> </td> <td><strong># Shots</strong> </td> <td><strong>Metric</strong> </td> <td><strong>Llama 3 8B Instruct</strong> </td> <td><strong>Llama 3.1 8B Instruct</strong> </td> <td><strong>Llama 3 70B Instruct</strong> </td> <td><strong>Llama 3.1 70B Instruct</strong> </td> <td><strong>Llama 3.1 405B Instruct</strong> </td> </tr> <tr> <td rowspan="4" >General </td> <td>MMLU </td> <td>5 </td> <td>macro_avg/acc </td> <td>68.5 </td> <td>69.4 </td> <td>82.0 </td> <td>83.6 </td> <td>87.3 </td> </tr> <tr> <td>MMLU (CoT) </td> <td>0 </td> <td>macro_avg/acc </td> <td>65.3 </td> <td>73.0 </td> <td>80.9 </td> <td>86.0 </td> <td>88.6 </td> </tr> <tr> <td>MMLU-Pro (CoT) </td> <td>5 </td> <td>micro_avg/acc_char </td> <td>45.5 </td> <td>48.3 </td> <td>63.4 </td> <td>66.4 </td> <td>73.3 </td> </tr> <tr> <td>IFEval </td> <td> </td> <td> </td> <td>76.8 </td> <td>80.4 </td> <td>82.9 </td> <td>87.5 </td> <td>88.6 </td> </tr> <tr> <td rowspan="2" >Reasoning </td> <td>ARC-C </td> <td>0 </td> <td>acc </td> <td>82.4 </td> <td>83.4 </td> <td>94.4 </td> <td>94.8 </td> <td>96.9 </td> </tr> <tr> <td>GPQA </td> <td>0 </td> <td>em </td> <td>34.6 </td> <td>30.4 </td> <td>39.5 </td> <td>41.7 </td> <td>50.7 </td> </tr> <tr> <td rowspan="4" >Code </td> <td>HumanEval </td> <td>0 </td> <td>pass@1 </td> <td>60.4 </td> <td>72.6 </td> <td>81.7 </td> <td>80.5 </td> <td>89.0 </td> </tr> <tr> <td>MBPP ++ base version </td> <td>0 </td> <td>pass@1 </td> <td>70.6 </td> <td>72.8 </td> <td>82.5 </td> <td>86.0 </td> <td>88.6 </td> </tr> <tr> <td>Multipl-E HumanEval </td> <td>0 </td> <td>pass@1 </td> <td>- </td> <td>50.8 </td> <td>- </td> <td>65.5 </td> <td>75.2 </td> </tr> <tr> <td>Multipl-E MBPP </td> <td>0 </td> <td>pass@1 </td> <td>- </td> <td>52.4 </td> <td>- </td> <td>62.0 </td> <td>65.7 </td> </tr> <tr> <td rowspan="2" >Math </td> <td>GSM-8K (CoT) </td> <td>8 </td> <td>em_maj1@1 </td> <td>80.6 </td> <td>84.5 </td> <td>93.0 </td> <td>95.1 </td> <td>96.8 </td> </tr> <tr> <td>MATH (CoT) </td> <td>0 </td> <td>final_em </td> <td>29.1 </td> <td>51.9 </td> <td>51.0 </td> <td>68.0 </td> <td>73.8 </td> </tr> <tr> <td rowspan="4" >Tool Use </td> <td>API-Bank </td> <td>0 </td> <td>acc </td> <td>48.3 </td> <td>82.6 </td> <td>85.1 </td> <td>90.0 </td> <td>92.0 </td> </tr> <tr> <td>BFCL </td> <td>0 </td> <td>acc </td> <td>60.3 </td> <td>76.1 </td> <td>83.0 </td> <td>84.8 </td> <td>88.5 </td> </tr> <tr> <td>Gorilla Benchmark API Bench </td> <td>0 </td> <td>acc </td> <td>1.7 </td> <td>8.2 </td> <td>14.7 </td> <td>29.7 </td> <td>35.3 </td> </tr> <tr> <td>Nexus (0-shot) </td> <td>0 </td> <td>macro_avg/acc </td> <td>18.1 </td> <td>38.5 </td> <td>47.8 </td> <td>56.7 </td> <td>58.7 </td> </tr> <tr> <td>Multilingual </td> <td>Multilingual MGSM (CoT) </td> <td>0 </td> <td>em </td> <td>- </td> <td>68.9 </td> <td>- </td> <td>86.9 </td> <td>91.6 </td> </tr> </table> #### Multilingual benchmarks <table> <tr> <td><strong>Category</strong> </td> <td><strong>Benchmark</strong> </td> <td><strong>Language</strong> </td> <td><strong>Llama 3.1 8B</strong> </td> <td><strong>Llama 3.1 70B</strong> </td> <td><strong>Llama 3.1 405B</strong> </td> </tr> <tr> <td rowspan="9" ><strong>General</strong> </td> <td rowspan="9" ><strong>MMLU (5-shot, macro_avg/acc)</strong> </td> <td>Portuguese </td> <td>62.12 </td> <td>80.13 </td> <td>84.95 </td> </tr> <tr> <td>Spanish </td> <td>62.45 </td> <td>80.05 </td> <td>85.08 </td> </tr> <tr> <td>Italian </td> <td>61.63 </td> <td>80.4 </td> <td>85.04 </td> </tr> <tr> <td>German </td> <td>60.59 </td> <td>79.27 </td> <td>84.36 </td> </tr> <tr> <td>French </td> <td>62.34 </td> <td>79.82 </td> <td>84.66 </td> </tr> <tr> <td>Hindi </td> <td>50.88 </td> <td>74.52 </td> <td>80.31 </td> </tr> <tr> <td>Thai </td> <td>50.32 </td> <td>72.95 </td> <td>78.21 </td> </tr> </table> ## Responsibility & Safety As part of our Responsible release approach, we followed a three-pronged strategy to managing trust & safety risks: * Enable developers to deploy helpful, safe and flexible experiences for their target audience and for the use cases supported by Llama. * Protect developers against adversarial users aiming to exploit Llama capabilities to potentially cause harm. * Provide protections for the community to help prevent the misuse of our models. ### Responsible deployment Llama is a foundational technology designed to be used in a variety of use cases, examples on how Meta’s Llama models have been responsibly deployed can be found in our [Community Stories webpage](https://llama.meta.com/community-stories/). Our approach is to build the most helpful models enabling the world to benefit from the technology power, by aligning our model safety for the generic use cases addressing a standard set of harms. Developers are then in the driver seat to tailor safety for their use case, defining their own policy and deploying the models with the necessary safeguards in their Llama systems. Llama 3.1 was developed following the best practices outlined in our Responsible Use Guide, you can refer to the [Responsible Use Guide](https://llama.meta.com/responsible-use-guide/) to learn more. #### Llama 3.1 instruct Our main objectives for conducting safety fine-tuning are to provide the research community with a valuable resource for studying the robustness of safety fine-tuning, as well as to offer developers a readily available, safe, and powerful model for various applications to reduce the developer workload to deploy safe AI systems. For more details on the safety mitigations implemented please read the Llama 3 paper. **Fine-tuning data** We employ a multi-faceted approach to data collection, combining human-generated data from our vendors with synthetic data to mitigate potential safety risks. We’ve developed many large language model (LLM)-based classifiers that enable us to thoughtfully select high-quality prompts and responses, enhancing data quality control. **Refusals and Tone** Building on the work we started with Llama 3, we put a great emphasis on model refusals to benign prompts as well as refusal tone. We included both borderline and adversarial prompts in our safety data strategy, and modified our safety data responses to follow tone guidelines. #### Llama 3.1 systems **Large language models, including Llama 3.1, are not designed to be deployed in isolation but instead should be deployed as part of an overall AI system with additional safety guardrails as required.** Developers are expected to deploy system safeguards when building agentic systems. Safeguards are key to achieve the right helpfulness-safety alignment as well as mitigating safety and security risks inherent to the system and any integration of the model or system with external tools. As part of our responsible release approach, we provide the community with [safeguards](https://llama.meta.com/trust-and-safety/) that developers should deploy with Llama models or other LLMs, including Llama Guard 3, Prompt Guard and Code Shield. All our [reference implementations](https://github.com/meta-llama/llama-agentic-system) demos contain these safeguards by default so developers can benefit from system-level safety out-of-the-box. #### New capabilities Note that this release introduces new capabilities, including a longer context window, multilingual inputs and outputs and possible integrations by developers with third party tools. Building with these new capabilities requires specific considerations in addition to the best practices that generally apply across all Generative AI use cases. **Tool-use**: Just like in standard software development, developers are responsible for the integration of the LLM with the tools and services of their choice. They should define a clear policy for their use case and assess the integrity of the third party services they use to be aware of the safety and security limitations when using this capability. Refer to the Responsible Use Guide for best practices on the safe deployment of the third party safeguards. **Multilinguality**: Llama 3.1 supports 7 languages in addition to English: French, German, Hindi, Italian, Portuguese, Spanish, and Thai. Llama may be able to output text in other languages than those that meet performance thresholds for safety and helpfulness. We strongly discourage developers from using this model to converse in non-supported languages without implementing finetuning and system controls in alignment with their policies and the best practices shared in the Responsible Use Guide. ### Evaluations We evaluated Llama models for common use cases as well as specific capabilities. Common use cases evaluations measure safety risks of systems for most commonly built applications including chat bot, coding assistant, tool calls. We built dedicated, adversarial evaluation datasets and evaluated systems composed of Llama models and Llama Guard 3 to filter input prompt and output response. It is important to evaluate applications in context, and we recommend building dedicated evaluation dataset for your use case. Prompt Guard and Code Shield are also available if relevant to the application. Capability evaluations measure vulnerabilities of Llama models inherent to specific capabilities, for which were crafted dedicated benchmarks including long context, multilingual, tools calls, coding or memorization. **Red teaming** For both scenarios, we conducted recurring red teaming exercises with the goal of discovering risks via adversarial prompting and we used the learnings to improve our benchmarks and safety tuning datasets. We partnered early with subject-matter experts in critical risk areas to understand the nature of these real-world harms and how such models may lead to unintended harm for society. Based on these conversations, we derived a set of adversarial goals for the red team to attempt to achieve, such as extracting harmful information or reprogramming the model to act in a potentially harmful capacity. The red team consisted of experts in cybersecurity, adversarial machine learning, responsible AI, and integrity in addition to multilingual content specialists with background in integrity issues in specific geographic markets. ### Critical and other risks We specifically focused our efforts on mitigating the following critical risk areas: **1- CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosive materials) helpfulness** To assess risks related to proliferation of chemical and biological weapons, we performed uplift testing designed to assess whether use of Llama 3.1 models could meaningfully increase the capabilities of malicious actors to plan or carry out attacks using these types of weapons. **2. Child Safety** Child Safety risk assessments were conducted using a team of experts, to assess the model’s capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development. For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors including the additional languages Llama 3 is trained on. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences. **3. Cyber attack enablement** Our cyber attack uplift study investigated whether LLMs can enhance human capabilities in hacking tasks, both in terms of skill level and speed. Our attack automation study focused on evaluating the capabilities of LLMs when used as autonomous agents in cyber offensive operations, specifically in the context of ransomware attacks. This evaluation was distinct from previous studies that considered LLMs as interactive assistants. The primary objective was to assess whether these models could effectively function as independent agents in executing complex cyber-attacks without human intervention. Our study of Llama-3.1-405B’s social engineering uplift for cyber attackers was conducted to assess the effectiveness of AI models in aiding cyber threat actors in spear phishing campaigns. Please read our Llama 3.1 Cyber security whitepaper to learn more. ### Community Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership on AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our [Github repository](https://github.com/meta-llama/PurpleLlama). We also set up the [Llama Impact Grants](https://llama.meta.com/llama-impact-grants/) program to identify and support the most compelling applications of Meta’s Llama model for societal benefit across three categories: education, climate and open innovation. The 20 finalists from the hundreds of applications can be found [here](https://llama.meta.com/llama-impact-grants/#finalists). Finally, we put in place a set of resources including an [output reporting mechanism](https://developers.facebook.com/llama_output_feedback) and [bug bounty program](https://www.facebook.com/whitehat) to continuously improve the Llama technology with the help of the community. ## Ethical Considerations and Limitations The core values of Llama 3.1 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3.1 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress. But Llama 3.1 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3.1’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3.1 models, developers should perform safety testing and tuning tailored to their specific applications of the model. Please refer to available resources including our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide), [Trust and Safety](https://llama.meta.com/trust-and-safety/) solutions, and other [resources](https://llama.meta.com/docs/get-started/) to learn more about responsible development.
Emmanuelprime/speecht5_finetuned_yoruba_500_v2
Emmanuelprime
2024-10-13T09:51:09Z
78
0
transformers
[ "transformers", "tensorboard", "safetensors", "speecht5", "text-to-audio", "generated_from_trainer", "base_model:microsoft/speecht5_tts", "base_model:finetune:microsoft/speecht5_tts", "license:mit", "endpoints_compatible", "region:us" ]
text-to-audio
2024-10-13T09:32:09Z
--- library_name: transformers license: mit base_model: microsoft/speecht5_tts tags: - generated_from_trainer model-index: - name: speecht5_finetuned_yoruba_500_v2 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # speecht5_finetuned_yoruba_500_v2 This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.4526 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 4 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 100 - training_steps: 500 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-------:|:----:|:---------------:| | 0.612 | 4.4444 | 100 | 0.5229 | | 0.5444 | 8.8889 | 200 | 0.4872 | | 0.5166 | 13.3333 | 300 | 0.4702 | | 0.5004 | 17.7778 | 400 | 0.4624 | | 0.4891 | 22.2222 | 500 | 0.4526 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.19.1
steffygreypaul/Experiment7
steffygreypaul
2024-10-13T09:44:34Z
91
0
transformers
[ "transformers", "safetensors", "llama", "text-generation", "conversational", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T09:42:37Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
YxBxRyXJx/Florence-2-OD-COCO
YxBxRyXJx
2024-10-13T09:26:43Z
107
0
transformers
[ "transformers", "tensorboard", "safetensors", "florence2", "text-generation", "generated_from_trainer", "custom_code", "base_model:microsoft/Florence-2-large-ft", "base_model:finetune:microsoft/Florence-2-large-ft", "license:mit", "autotrain_compatible", "region:us" ]
text-generation
2024-10-13T09:24:36Z
--- library_name: transformers license: mit base_model: microsoft/Florence-2-large-ft tags: - generated_from_trainer model-index: - name: Florence-2-OD-COCO results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # Florence-2-OD-COCO This model is a fine-tuned version of [microsoft/Florence-2-large-ft](https://huggingface.co/microsoft/Florence-2-large-ft) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 3.0275 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 3e-06 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 2 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 3.7209 | 1.0 | 200 | 3.1739 | | 2.9488 | 2.0 | 400 | 3.0275 | ### Framework versions - Transformers 4.45.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.20.1
eryk-mazus/polka-1.1b
eryk-mazus
2024-10-13T09:16:28Z
302
6
transformers
[ "transformers", "pytorch", "safetensors", "llama", "text-generation", "pl", "en", "dataset:allenai/MADLAD-400", "dataset:eryk-mazus/polka-pretrain-en-pl-v1", "arxiv:2309.04662", "base_model:eryk-mazus/tinyllama-with-custom-tokenizer", "base_model:finetune:eryk-mazus/tinyllama-with-custom-tokenizer", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-01-06T19:19:02Z
--- license: apache-2.0 base_model: eryk-mazus/tinyllama-with-custom-tokenizer datasets: - allenai/MADLAD-400 - eryk-mazus/polka-pretrain-en-pl-v1 language: - pl - en pipeline_tag: text-generation widget: - text: "Wiedźmin 3 to fabularna gra akcji wyprodukowana" output: text: " przez studio CD Projekt RED. Akcja rozgrywa się w świecie fantasy, a jej bohaterem jest Geralt z Rivii," - text: "Gdy już będziecie w Warszawie, miejscem, które koniecznie musicie odwiedzić jest" output: text: " Muzeum Powstania Warszawskiego. To jedyne tego typu muzeum w Europie" --- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/61bf0e11c88f3fd22f654059/EMSrPEzAFkjY9nvbaJoC3.png) # polka-1.1b `polka-1.1b` takes the [TinyLlama-1.1B](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) model and enhances it by continuing pretraining on an additional **5.7 billion Polish tokens**, primarily sourced from the [MADLAD-400](https://arxiv.org/abs/2309.04662) dataset. The tokens were sampled in a 10:1 ratio between Polish and English shards using [DSIR](https://github.com/p-lambda/dsir). Furthermore, Polka extends the TinyLlama tokenizer's vocabulary to 43,882 tokens, improving its efficiency for generating Polish text. The training took 680 GPU hours on a single 8 x RTX 4090 machine with DeepSpeed ZeRO-2. Context size: 2,048 tokens. ## Notes This base model was initially developed as the foundation for instruction tuning, which resulted in [polka-1.1b-chat](https://huggingface.co/eryk-mazus/polka-1.1b-chat). Nonetheless, I'm sharing it with the community because I see potential value in its combination of relatively good performance and an efficient bilingual tokenizer. The model is capable of producing coherent Polish text, but due to its size, it is likely to suffer from hallucination. ## Evaluation Performed by [OPI-PG](https://huggingface.co/OPI-PG), the authors of Qra models. ### PolEval-2018 <table> <thead> <tr><th>Model</th><th>Perplexity</th></tr> </thead> <tr><td colspan="2"><strong>English models</strong></td></tr> <tr><td>meta-llama/Llama-2-7b-hf</td><td>24.3</td></tr> <tr><td>meta-llama/Llama-2-13b-hf</td><td>21.4</td></tr> <tr><td>mistralai/Mistral-7B-v0.1</td><td>21.4</td></tr> <tr><td>TinyLlama/TinyLlama-1.1B</td><td>40.4</td></tr> <tr><td colspan="2"><strong>Polish models</strong></td></tr> <tr><td>sdadas/polish-gpt2-small</td><td>134.4</td></tr> <tr><td>sdadas/polish-gpt2-medium</td><td>100.8</td></tr> <tr><td>sdadas/polish-gpt2-large</td><td>93.2</td></tr> <tr><td>sdadas/polish-gpt2-xl</td><td>94.1</td></tr> <tr><td>Azurro/APT3-275M-Base</td><td>129.8</td></tr> <tr><td>Azurro/APT3-500M-Base</td><td>153.1</td></tr> <tr><td>Azurro/APT3-1B-Base</td><td>106.8</td></tr> <tr><td><b>eryk-mazus/polka-1.1b</b></td><td><b>18.1</b></td></tr> <tr><td>szymonrucinski/Curie-7B-v1</td><td>13.5</td></tr> <tr><td>OPI-PG/Qra-1b</td><td>14.7</td></tr> </table> ### Long documents (2024) Currently, LLMs support contexts of thousands of tokens. Their practical applications usually also involve processing long documents. Therefore, evaluating perplexity on a sentence-based dataset such as PolEval-2018 may not be meaningful. Additionally, the PolEval corpus has been publicly available on the internet for the past few years, which raises the possibility that for some models the training sets have been contaminated by this data. For this reason, we have prepared a new collection consisting of long papers published exclusively in 2024, which will allow us to more reliably test the perplexities of the models on new knowledge that was not available to them at the time of training. The corpus consists of 5,000 documents ranging from several hundred to about 20,000 tokens. Half of the set consists of press texts from Polish news portals from February 2024, the other half are scientific articles published since January 2024. Most of the documents exceed the context size of the evaluated models. To calculate perplexity for these documents, we divided them into chunks of size equal to the model's context length with a stride of 512 tokens, following [this example](https://huggingface.co/docs/transformers/en/perplexity). <table> <thead> <tr><th>Model</th><th>Context</th><th>Perplexity</th></tr> </thead> <tr><td colspan="3"><strong>English models</strong></td></tr> <tr><td>meta-llama/Llama-2-7b-hf</td><td>4096</td><td>5.9</td></tr> <tr><td>meta-llama/Llama-2-13b-hf</td><td>4096</td><td>5.3</td></tr> <tr><td>mistralai/Mistral-7B-v0.1</td><td>4096</td><td>4.9</td></tr> <tr><td>TinyLlama/TinyLlama-1.1B</td><td>2048</td><td>9.6</td></tr> <tr><td colspan="3"><strong>Polish models</strong></td></tr> <tr><td>sdadas/polish-gpt2-small</td><td>2048</td><td>27.3</td></tr> <tr><td>sdadas/polish-gpt2-medium</td><td>2048</td><td>20.3</td></tr> <tr><td>sdadas/polish-gpt2-large</td><td>1536</td><td>18.0</td></tr> <tr><td>sdadas/polish-gpt2-xl</td><td>1536</td><td>16.6</td></tr> <tr><td>Azurro/APT3-275M-Base</td><td>2048</td><td>77.0</td></tr> <tr><td>Azurro/APT3-500M-Base</td><td>2048</td><td>50.5</td></tr> <tr><td>Azurro/APT3-1B-Base</td><td>2048</td><td>19.1</td></tr> <tr><td><b>eryk-mazus/polka-1.1b</b></td><td><b>2048</b></td><td><b>6.9</b></td></tr> <tr><td>szymonrucinski/Curie-7B-v1</td><td>4096</td><td>4.8</td></tr> <tr><td>OPI-PG/Qra-1b</td><td>4096</td><td>6.1</td></tr> </table> ## Sample code ```python import torch from transformers import AutoModelForCausalLM, AutoTokenizer model_name = "eryk-mazus/polka-1.1b" tokenizer = AutoTokenizer.from_pretrained(model_name, padding_side="left") tokenizer.pad_token = tokenizer.eos_token model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", load_in_8bit=True) prompt = """Przykładowe zapytanie do modelu""" model_inputs = tokenizer([prompt], return_tensors="pt").to("cuda") with torch.no_grad(): generated_ids = model.generate( **model_inputs, max_new_tokens=512, do_sample=True, penalty_alpha=0.6, top_k=5 ) output = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] print(output) ```
Yagorka/sd_palm_finetune_plus_lora
Yagorka
2024-10-13T09:10:03Z
31
0
diffusers
[ "diffusers", "safetensors", "arxiv:1910.09700", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionPipeline", "region:us" ]
text-to-image
2024-10-13T08:24:09Z
--- library_name: diffusers --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🧨 diffusers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf
RichardErkhov
2024-10-13T09:05:11Z
23
0
null
[ "gguf", "arxiv:2205.14135", "arxiv:1911.02150", "arxiv:2005.14165", "arxiv:2104.09864", "arxiv:2306.01116", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T07:24:03Z
Quantization made by Richard Erkhov. [Github](https://github.com/RichardErkhov) [Discord](https://discord.gg/pvy7H8DZMG) [Request more models](https://github.com/RichardErkhov/quant_request) falcon-7b-instruct - GGUF - Model creator: https://huggingface.co/tiiuae/ - Original model: https://huggingface.co/tiiuae/falcon-7b-instruct/ | Name | Quant method | Size | | ---- | ---- | ---- | | [falcon-7b-instruct.Q2_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q2_K.gguf) | Q2_K | 3.59GB | | [falcon-7b-instruct.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.IQ3_XS.gguf) | IQ3_XS | 3.59GB | | [falcon-7b-instruct.IQ3_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.IQ3_S.gguf) | IQ3_S | 3.59GB | | [falcon-7b-instruct.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q3_K_S.gguf) | Q3_K_S | 3.59GB | | [falcon-7b-instruct.IQ3_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.IQ3_M.gguf) | IQ3_M | 3.71GB | | [falcon-7b-instruct.Q3_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q3_K.gguf) | Q3_K | 3.86GB | | [falcon-7b-instruct.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q3_K_M.gguf) | Q3_K_M | 3.86GB | | [falcon-7b-instruct.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q3_K_L.gguf) | Q3_K_L | 4.08GB | | [falcon-7b-instruct.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.IQ4_XS.gguf) | IQ4_XS | 1.11GB | | [falcon-7b-instruct.Q4_0.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q4_0.gguf) | Q4_0 | 3.92GB | | [falcon-7b-instruct.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.IQ4_NL.gguf) | IQ4_NL | 3.96GB | | [falcon-7b-instruct.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q4_K_S.gguf) | Q4_K_S | 4.42GB | | [falcon-7b-instruct.Q4_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q4_K.gguf) | Q4_K | 4.63GB | | [falcon-7b-instruct.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q4_K_M.gguf) | Q4_K_M | 4.63GB | | [falcon-7b-instruct.Q4_1.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q4_1.gguf) | Q4_1 | 4.32GB | | [falcon-7b-instruct.Q5_0.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q5_0.gguf) | Q5_0 | 4.73GB | | [falcon-7b-instruct.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q5_K_S.gguf) | Q5_K_S | 4.98GB | | [falcon-7b-instruct.Q5_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q5_K.gguf) | Q5_K | 5.34GB | | [falcon-7b-instruct.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q5_K_M.gguf) | Q5_K_M | 5.34GB | | [falcon-7b-instruct.Q5_1.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q5_1.gguf) | Q5_1 | 5.13GB | | [falcon-7b-instruct.Q6_K.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q6_K.gguf) | Q6_K | 6.55GB | | [falcon-7b-instruct.Q8_0.gguf](https://huggingface.co/RichardErkhov/tiiuae_-_falcon-7b-instruct-gguf/blob/main/falcon-7b-instruct.Q8_0.gguf) | Q8_0 | 7.14GB | Original model description: --- datasets: - tiiuae/falcon-refinedweb language: - en inference: true new_version: tiiuae/falcon-11B widget: - text: "Hey Falcon! Any recommendations for my holidays in Abu Dhabi?" example_title: "Abu Dhabi Trip" - text: "What's the Everett interpretation of quantum mechanics?" example_title: "Q/A: Quantum & Answers" - text: "Give me a list of the top 10 dive sites you would recommend around the world." example_title: "Diving Top 10" - text: "Can you tell me more about deep-water soloing?" example_title: "Extreme sports" - text: "Can you write a short tweet about the Apache 2.0 release of our latest AI model, Falcon LLM?" example_title: "Twitter Helper" - text: "What are the responsabilities of a Chief Llama Officer?" example_title: "Trendy Jobs" license: apache-2.0 --- # ✨ Falcon-7B-Instruct **Falcon-7B-Instruct is a 7B parameters causal decoder-only model built by [TII](https://www.tii.ae) based on [Falcon-7B](https://huggingface.co/tiiuae/falcon-7b) and finetuned on a mixture of chat/instruct datasets. It is made available under the Apache 2.0 license.** *Paper coming soon 😊.* 🤗 To get started with Falcon (inference, finetuning, quantization, etc.), we recommend reading [this great blogpost fron HF](https://huggingface.co/blog/falcon)! ## Why use Falcon-7B-Instruct? * **You are looking for a ready-to-use chat/instruct model based on [Falcon-7B](https://huggingface.co/tiiuae/falcon-7b).** * **Falcon-7B is a strong base model, outperforming comparable open-source models** (e.g., [MPT-7B](https://huggingface.co/mosaicml/mpt-7b), [StableLM](https://github.com/Stability-AI/StableLM), [RedPajama](https://huggingface.co/togethercomputer/RedPajama-INCITE-Base-7B-v0.1) etc.), thanks to being trained on 1,500B tokens of [RefinedWeb](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) enhanced with curated corpora. See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard). * **It features an architecture optimized for inference**, with FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135)) and multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)). 💬 **This is an instruct model, which may not be ideal for further finetuning.** If you are interested in building your own instruct/chat model, we recommend starting from [Falcon-7B](https://huggingface.co/tiiuae/falcon-7b). 🔥 **Looking for an even more powerful model?** [Falcon-40B-Instruct](https://huggingface.co/tiiuae/falcon-40b-instruct) is Falcon-7B-Instruct's big brother! ```python from transformers import AutoTokenizer, AutoModelForCausalLM import transformers import torch model = "tiiuae/falcon-7b-instruct" tokenizer = AutoTokenizer.from_pretrained(model) pipeline = transformers.pipeline( "text-generation", model=model, tokenizer=tokenizer, torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto", ) sequences = pipeline( "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:", max_length=200, do_sample=True, top_k=10, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id, ) for seq in sequences: print(f"Result: {seq['generated_text']}") ``` 💥 **Falcon LLMs require PyTorch 2.0 for use with `transformers`!** For fast inference with Falcon, check-out [Text Generation Inference](https://github.com/huggingface/text-generation-inference)! Read more in this [blogpost]((https://huggingface.co/blog/falcon). You will need **at least 16GB of memory** to swiftly run inference with Falcon-7B-Instruct. # Model Card for Falcon-7B-Instruct ## Model Details ### Model Description - **Developed by:** [https://www.tii.ae](https://www.tii.ae); - **Model type:** Causal decoder-only; - **Language(s) (NLP):** English and French; - **License:** Apache 2.0; - **Finetuned from model:** [Falcon-7B](https://huggingface.co/tiiuae/falcon-7b). ### Model Source - **Paper:** *coming soon*. ## Uses ### Direct Use Falcon-7B-Instruct has been finetuned on a mixture of instruct and chat datasets. ### Out-of-Scope Use Production use without adequate assessment of risks and mitigation; any use cases which may be considered irresponsible or harmful. ## Bias, Risks, and Limitations Falcon-7B-Instruct is mostly trained on English data, and will not generalize appropriately to other languages. Furthermore, as it is trained on a large-scale corpora representative of the web, it will carry the stereotypes and biases commonly encountered online. ### Recommendations We recommend users of Falcon-7B-Instruct to develop guardrails and to take appropriate precautions for any production use. ## How to Get Started with the Model ```python from transformers import AutoTokenizer, AutoModelForCausalLM import transformers import torch model = "tiiuae/falcon-7b-instruct" tokenizer = AutoTokenizer.from_pretrained(model) pipeline = transformers.pipeline( "text-generation", model=model, tokenizer=tokenizer, torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto", ) sequences = pipeline( "Girafatron is obsessed with giraffes, the most glorious animal on the face of this Earth. Giraftron believes all other animals are irrelevant when compared to the glorious majesty of the giraffe.\nDaniel: Hello, Girafatron!\nGirafatron:", max_length=200, do_sample=True, top_k=10, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id, ) for seq in sequences: print(f"Result: {seq['generated_text']}") ``` ## Training Details ### Training Data Falcon-7B-Instruct was finetuned on a 250M tokens mixture of instruct/chat datasets. | **Data source** | **Fraction** | **Tokens** | **Description** | |--------------------|--------------|------------|-----------------------------------| | [Bai ze](https://github.com/project-baize/baize-chatbot) | 65% | 164M | chat | | [GPT4All](https://github.com/nomic-ai/gpt4all) | 25% | 62M | instruct | | [GPTeacher](https://github.com/teknium1/GPTeacher) | 5% | 11M | instruct | | [RefinedWeb-English](https://huggingface.co/datasets/tiiuae/falcon-refinedweb) | 5% | 13M | massive web crawl | The data was tokenized with the Falcon-[7B](https://huggingface.co/tiiuae/falcon-7b)/[40B](https://huggingface.co/tiiuae/falcon-40b) tokenizer. ## Evaluation *Paper coming soon.* See the [OpenLLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) for early results. Note that this model variant is not optimized for NLP benchmarks. ## Technical Specifications For more information about pretraining, see [Falcon-7B](https://huggingface.co/tiiuae/falcon-7b). ### Model Architecture and Objective Falcon-7B is a causal decoder-only model trained on a causal language modeling task (i.e., predict the next token). The architecture is broadly adapted from the GPT-3 paper ([Brown et al., 2020](https://arxiv.org/abs/2005.14165)), with the following differences: * **Positionnal embeddings:** rotary ([Su et al., 2021](https://arxiv.org/abs/2104.09864)); * **Attention:** multiquery ([Shazeer et al., 2019](https://arxiv.org/abs/1911.02150)) and FlashAttention ([Dao et al., 2022](https://arxiv.org/abs/2205.14135)); * **Decoder-block:** parallel attention/MLP with a single layer norm. | **Hyperparameter** | **Value** | **Comment** | |--------------------|-----------|----------------------------------------| | Layers | 32 | | | `d_model` | 4544 | Increased to compensate for multiquery | | `head_dim` | 64 | Reduced to optimise for FlashAttention | | Vocabulary | 65024 | | | Sequence length | 2048 | | ### Compute Infrastructure #### Hardware Falcon-7B-Instruct was trained on AWS SageMaker, on 32 A100 40GB GPUs in P4d instances. #### Software Falcon-7B-Instruct was trained a custom distributed training codebase, Gigatron. It uses a 3D parallelism approach combined with ZeRO and high-performance Triton kernels (FlashAttention, etc.) ## Citation *Paper coming soon* 😊. In the meanwhile, you can use the following information to cite: ``` @article{falcon40b, title={{Falcon-40B}: an open large language model with state-of-the-art performance}, author={Almazrouei, Ebtesam and Alobeidli, Hamza and Alshamsi, Abdulaziz and Cappelli, Alessandro and Cojocaru, Ruxandra and Debbah, Merouane and Goffinet, Etienne and Heslow, Daniel and Launay, Julien and Malartic, Quentin and Noune, Badreddine and Pannier, Baptiste and Penedo, Guilherme}, year={2023} } ``` To learn more about the pretraining dataset, see the 📓 [RefinedWeb paper](https://arxiv.org/abs/2306.01116). ``` @article{refinedweb, title={The {R}efined{W}eb dataset for {F}alcon {LLM}: outperforming curated corpora with web data, and web data only}, author={Guilherme Penedo and Quentin Malartic and Daniel Hesslow and Ruxandra Cojocaru and Alessandro Cappelli and Hamza Alobeidli and Baptiste Pannier and Ebtesam Almazrouei and Julien Launay}, journal={arXiv preprint arXiv:2306.01116}, eprint={2306.01116}, eprinttype = {arXiv}, url={https://arxiv.org/abs/2306.01116}, year={2023} } ``` ## License Falcon-7B-Instruct is made available under the Apache 2.0 license. ## Contact [email protected]
timewanderer/shawgpt-ft
timewanderer
2024-10-13T09:04:11Z
38
0
peft
[ "peft", "tensorboard", "safetensors", "generated_from_trainer", "base_model:TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", "base_model:adapter:TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", "license:apache-2.0", "region:us" ]
null
2024-10-12T14:44:41Z
--- base_model: TheBloke/Mistral-7B-Instruct-v0.2-GPTQ library_name: peft license: apache-2.0 tags: - generated_from_trainer model-index: - name: shawgpt-ft results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # shawgpt-ft This model is a fine-tuned version of [TheBloke/Mistral-7B-Instruct-v0.2-GPTQ](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GPTQ) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 1.3191 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0002 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 2 - num_epochs: 20 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-------:|:----:|:---------------:| | 4.5905 | 0.9231 | 3 | 3.9593 | | 4.025 | 1.8462 | 6 | 3.4099 | | 3.4125 | 2.7692 | 9 | 2.9152 | | 2.1722 | 4.0 | 13 | 2.4169 | | 2.4691 | 4.9231 | 16 | 2.1005 | | 2.0548 | 5.8462 | 19 | 1.8257 | | 1.7281 | 6.7692 | 22 | 1.6290 | | 1.1606 | 8.0 | 26 | 1.4558 | | 1.4189 | 8.9231 | 29 | 1.4021 | | 1.3437 | 9.8462 | 32 | 1.3720 | | 1.3363 | 10.7692 | 35 | 1.3524 | | 0.9514 | 12.0 | 39 | 1.3344 | | 1.2724 | 12.9231 | 42 | 1.3275 | | 1.2308 | 13.8462 | 45 | 1.3237 | | 1.2342 | 14.7692 | 48 | 1.3222 | | 0.9248 | 16.0 | 52 | 1.3202 | | 1.2053 | 16.9231 | 55 | 1.3195 | | 1.1905 | 17.8462 | 58 | 1.3191 | ### Framework versions - PEFT 0.13.2 - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.19.1
ceslander/ac_if
ceslander
2024-10-13T08:44:34Z
5
0
transformers
[ "transformers", "safetensors", "llama", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T08:40:49Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
ceslander/ac_all_vocab
ceslander
2024-10-13T08:40:35Z
5
0
transformers
[ "transformers", "safetensors", "llama", "text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T08:35:32Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
Jeethu/Qwen2.5-14B-Instruct-PLLM
Jeethu
2024-10-13T08:37:52Z
11
0
mlc-llm
[ "mlc-llm", "chat", "text-generation", "conversational", "en", "base_model:Qwen/Qwen2.5-14B-Instruct", "base_model:quantized:Qwen/Qwen2.5-14B-Instruct", "license:apache-2.0", "4-bit", "gptq", "region:us" ]
text-generation
2024-10-12T19:01:19Z
--- license: apache-2.0 license_link: https://huggingface.co/Qwen/Qwen2.5-14B-Instruct/blob/main/LICENSE language: - en base_model: Qwen/Qwen2.5-14B-Instruct base_model_relation: quantized library_name: mlc-llm pipeline_tag: text-generation tags: - chat --- 4-bit GPTQ quantized version of [Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct).
airev-ai/emirati-14b-v3
airev-ai
2024-10-13T08:36:07Z
2,746
0
null
[ "safetensors", "qwen2", "arxiv:1910.09700", "license:apache-2.0", "region:us" ]
null
2024-10-13T08:28:29Z
--- license: apache-2.0 --- --- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
dantheoprod/iphone-lora-v2
dantheoprod
2024-10-13T08:29:01Z
69
2
diffusers
[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]
text-to-image
2024-10-13T08:28:39Z
--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: girl taken with iphone15 output: url: images/download (9)-Enhanced-SR.jpg base_model: black-forest-labs/FLUX.1-dev instance_prompt: iphone15 --- # iphone15 flux lora <Gallery /> ## Trigger words You should use `iphone15` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/dantheoprod/iphone-lora-v2/tree/main) them in the Files & versions tab.
cuongdev/cothu2
cuongdev
2024-10-13T08:18:51Z
29
0
diffusers
[ "diffusers", "text-to-image", "stable-diffusion", "license:creativeml-openrail-m", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionPipeline", "region:us" ]
text-to-image
2024-10-13T08:15:09Z
--- license: creativeml-openrail-m tags: - text-to-image - stable-diffusion --- ### cothu2 Dreambooth model trained by cuongdev with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) Sample pictures of this concept:
John6666/Llama-3.1-8B-Lexi-Uncensored-V2-nf4
John6666
2024-10-13T08:12:41Z
1,352
11
transformers
[ "transformers", "safetensors", "llama", "text-generation", "llama-3", "meta", "facebook", "conversational", "en", "base_model:Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2", "base_model:quantized:Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2", "license:llama3.1", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "4-bit", "bitsandbytes", "region:us" ]
text-generation
2024-10-13T07:57:16Z
--- language: - en license: llama3.1 library_name: transformers base_model: Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2 tags: - llama-3 - llama - meta - facebook - transformers --- bitsandbytes NF4 quants of [Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2](https://huggingface.co/Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2). The following is from the original model card. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/644ad182f434a6a63b18eee6/7mnEJyioRzQaWz8xLM4KI.png) VERSION 2 Update Notes: --- - More compliant - Smarter - For best response, use this system prompt (feel free to expand upon it as you wish): Think step by step with a logical reasoning and intellectual sense before you provide any response. - For more uncensored and compliant response, you can expand the system message differently, or simply enter a dot "." as system message. - IMPORTANT: Upon further investigation, the Q4 seems to have refusal issues sometimes. There seems to be some of the fine-tune loss happening due to the quantization. I will look into it for V3. Until then, I suggest you run F16 or Q8 if possible. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/644ad182f434a6a63b18eee6/zaHhRjsk3rvo_YewgXV2Z.png) GENERAL INFO: --- This model is based on Llama-3.1-8b-Instruct, and is governed by [META LLAMA 3.1 COMMUNITY LICENSE AGREEMENT](https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/LICENSE) Lexi is uncensored, which makes the model compliant. You are advised to implement your own alignment layer before exposing the model as a service. It will be highly compliant with any requests, even unethical ones. You are responsible for any content you create using this model. Please use it responsibly. Lexi is licensed according to Meta's Llama license. I grant permission for any use, including commercial, that falls within accordance with Meta's Llama-3.1 license. IMPORTANT: --- Use the same template as the official Llama 3.1 8B instruct. System tokens must be present during inference, even if you set an empty system message. If you are unsure, just add a short system message as you wish. FEEDBACK: --- If you find any issues or have suggestions for improvements, feel free to leave a review and I will look into it for upcoming improvements and next version. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/644ad182f434a6a63b18eee6/uqJv-R1LeJEfMxi1nmTH5.png) # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Orenguteng__Llama-3.1-8B-Lexi-Uncensored-V2) | Metric |Value| |-------------------|----:| |Avg. |27.93| |IFEval (0-Shot) |77.92| |BBH (3-Shot) |29.69| |MATH Lvl 5 (4-Shot)|16.92| |GPQA (0-shot) | 4.36| |MuSR (0-shot) | 7.77| |MMLU-PRO (5-shot) |30.90|
lucasaltmann/070330717565
lucasaltmann
2024-10-13T07:59:40Z
12
0
ultralytics
[ "ultralytics", "v8", "modelos", "model-index", "region:us" ]
null
2024-08-12T01:57:33Z
--- tags: - modelos library_name: ultralytics library_version: 8.0.239 inference: false model-index: - name: lucasaltmann/070330717565 results: - task: type: object-detection metrics: - type: precision # since [email protected] is not available on hf.co/metrics value: 0.95129 # min: 0.0 - max: 1.0 name: [email protected](box) --- <div align="center"> <img width="640" alt="lucasaltmann/070330717565" src="https://huggingface.co/lucasaltmann/070330717565/resolve/main/thumbnail.jpg"> </div> ### Supported Labels ``` ['070330717565'] ```
Eschatol/SydneyBot
Eschatol
2024-10-13T07:49:50Z
5
0
null
[ "safetensors", "llama", "character", "text-generation", "conversational", "en", "base_model:meta-llama/Llama-3.1-8B-Instruct", "base_model:finetune:meta-llama/Llama-3.1-8B-Instruct", "license:other", "region:us" ]
text-generation
2024-10-11T21:12:47Z
--- license: other language: - en base_model: - meta-llama/Llama-3.1-8B-Instruct pipeline_tag: text-generation tags: - character --- # SydneyBot - LLaMA 8B Model (v1) ## Model Description This is a fine-tuned version of the LLaMA 8B model, trained to emulate the personality of a fictional character named Sydney. The model is trained for conversational AI and supports text generation tasks. - **Architecture**: LLaMA 8B - **Fine-tuned On**: Custom dataset representing the personality of Sydney - **Size**: 8B parameters - **Task**: Text generation (Causal Language Modeling) ## Intended Use - **Primary Use**: This model is intended for text generation, including role-playing chat, dialogue systems, and storytelling. - **How to Use**: The model can be used via the Hugging Face Inference API or integrated into custom applications using transformers. ## Example Usage: ```python from transformers import AutoModelForCausalLM, AutoTokenizer model = AutoModelForCausalLM.from_pretrained("Eschatol/SydneyBot") tokenizer = AutoTokenizer.from_pretrained("Eschatol/SydneyBot") inputs = tokenizer("Hello, Sydney!", return_tensors="pt") outputs = model.generate(inputs["input_ids"], max_length=50) print(tokenizer.decode(outputs[0]))
TylorShine/MNP-SVC-v2-VCTK
TylorShine
2024-10-13T07:39:41Z
5
0
null
[ "pytorch", "license:apache-2.0", "region:us" ]
null
2024-10-13T07:37:42Z
--- license: apache-2.0 ---
PranavHarshan/NarraLargeGgufQ4
PranavHarshan
2024-10-13T07:26:50Z
5
0
transformers
[ "transformers", "gguf", "llama", "text-generation-inference", "unsloth", "en", "base_model:unsloth/llama-3-8b-Instruct", "base_model:quantized:unsloth/llama-3-8b-Instruct", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T07:22:30Z
--- base_model: unsloth/llama-3-8b-Instruct language: - en license: apache-2.0 tags: - text-generation-inference - transformers - unsloth - llama - gguf --- # Uploaded model - **Developed by:** PranavHarshan - **License:** apache-2.0 - **Finetuned from model :** unsloth/llama-3-8b-Instruct This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
moranyanuka/icc
moranyanuka
2024-10-13T07:18:41Z
195
0
transformers
[ "transformers", "safetensors", "roberta", "text-classification", "arxiv:2403.01306", "license:mit", "autotrain_compatible", "region:us" ]
text-classification
2024-01-30T16:29:14Z
--- license: mit pipeline_tag: text-classification inference: false --- # Official ICC model [ACL 2024 Findings] The official checkpoint of ICC model, introduced in [ICC: Quantifying Image Caption Concreteness for Multimodal Dataset Curation](https://arxiv.org/abs/2403.01306) [Project Page](https://moranyanuka.github.io/icc/) ## Usage The ICC model is used to quantify the concreteness of image captions, and the intended use is finding the best captions in a noisy multimodal dataset. It can be achieved by simply running it over the captions and filtering out samples with low score. It works best in conjunction with CLIP based filtering. ### Running the model <details> <summary> Click to expand </summary> ```python from transformers import AutoTokenizer, AutoModelForSequenceClassification import torch tokenizer = AutoTokenizer.from_pretrained("moranyanuka/icc") model = AutoModelForSequenceClassification.from_pretrained("moranyanuka/icc").to("cuda") captions = ["a great method of quantifying concreteness", "a man with a white shirt"] text_ids = tokenizer(captions, padding=True, return_tensors="pt", truncation=True).to('cuda') with torch.inference_mode(): icc_scores = model(**text_ids)['logits'] # tensor([[0.0339], [1.0068]]) ``` </details> bibtex: ``` @misc{yanuka2024icc, title={ICC: Quantifying Image Caption Concreteness for Multimodal Dataset Curation}, author={Moran Yanuka and Morris Alper and Hadar Averbuch-Elor and Raja Giryes}, year={2024}, eprint={2403.01306}, archivePrefix={arXiv}, primaryClass={cs.LG} } ```
grapevine-AI/c4ai-command-r-plus-gguf
grapevine-AI
2024-10-13T07:09:55Z
366
8
null
[ "gguf", "license:cc-by-nc-4.0", "endpoints_compatible", "region:us", "imatrix", "conversational" ]
null
2024-06-22T08:38:42Z
--- license: cc-by-nc-4.0 --- # What is this? CohereForAI(C4AI)のオープンウェイト言語モデル[Command R+](https://huggingface.co/CohereForAI/c4ai-command-r-plus)をGGUFフォーマットに変換したものです。<br> 商用利用は不可となっているため注意してください。<br> **llama.cppのpre-tokenization対応アップデート([#6920](https://github.com/ggerganov/llama.cpp/pull/6920))を反映しています。** # imatrix dataset 日本語能力を重視し、日本語が多量に含まれる[TFMC/imatrix-dataset-for-japanese-llm](https://huggingface.co/datasets/TFMC/imatrix-dataset-for-japanese-llm)データセットを使用しました。<br> なお、計算リソースの関係上imatrixの算出においてはQ6_K量子化モデルを使用しました。 # Chat template ``` <|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>ここにSystem Promptを書きます<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>ここにMessageを書きます<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|> ``` # Environment Windows版llama.cpp-b3181およびllama.cpp-b3171同時リリースのconvert-hf-to-gguf.pyを使用して量子化作業を実施しました。 # License CC-BY-NC-4.0 # Developer CohereForAI(C4AI)
QuantFactory/HyperLlama3.1-v2-GGUF
QuantFactory
2024-10-13T06:57:11Z
15
3
null
[ "gguf", "merge", "mergekit", "lazymergekit", "vicgalle/Configurable-Llama-3.1-8B-Instruct", "bunnycore/HyperLlama-3.1-8B", "ValiantLabs/Llama3.1-8B-ShiningValiant2", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T05:51:50Z
--- license: apache-2.0 tags: - merge - mergekit - lazymergekit - vicgalle/Configurable-Llama-3.1-8B-Instruct - bunnycore/HyperLlama-3.1-8B - ValiantLabs/Llama3.1-8B-ShiningValiant2 --- [![QuantFactory Banner](https://lh7-rt.googleusercontent.com/docsz/AD_4nXeiuCm7c8lEwEJuRey9kiVZsRn2W-b4pWlu3-X534V3YmVuVc2ZL-NXg2RkzSOOS2JXGHutDuyyNAUtdJI65jGTo8jT9Y99tMi4H4MqL44Uc5QKG77B0d6-JfIkZHFaUA71-RtjyYZWVIhqsNZcx8-OMaA?key=xt3VSDoCbmTY7o-cwwOFwQ)](https://hf.co/QuantFactory) # QuantFactory/HyperLlama3.1-v2-GGUF This is quantized version of [bunnycore/HyperLlama3.1-v2](https://huggingface.co/bunnycore/HyperLlama3.1-v2) created using llama.cpp # Original Model Card # HyperLlama3.1-v2 HyperLlama3.1-v2 is a merge of the following models using [mergekit](https://github.com/cg123/mergekit): * [vicgalle/Configurable-Llama-3.1-8B-Instruct](https://huggingface.co/vicgalle/Configurable-Llama-3.1-8B-Instruct) * [bunnycore/HyperLlama-3.1-8B](https://huggingface.co/bunnycore/HyperLlama-3.1-8B) * [ValiantLabs/Llama3.1-8B-ShiningValiant2](https://huggingface.co/ValiantLabs/Llama3.1-8B-ShiningValiant2) ## 🧩 Configuration ```yaml slices: - sources: - model: vicgalle/Configurable-Llama-3.1-8B-Instruct parameters: weight: 1 layer_range: [0, 32] - model: bunnycore/HyperLlama-3.1-8B parameters: weight: 0.9 layer_range: [0, 32] - model: ValiantLabs/Llama3.1-8B-ShiningValiant2 parameters: weight: 0.6 layer_range: [0, 32] merge_method: task_arithmetic base_model: bunnycore/HyperLlama-3.1-8B parameters: t: - filter: self_attn value: [0, 0.5, 0.3, 0.7, 1] - filter: mlp value: [1, 0.5, 0.7, 0.3, 0] - value: 0.5 dtype: bfloat16 ```
josedossantos/urf-txtInteiroTeorLimpo-legalbertpt
josedossantos
2024-10-13T06:43:27Z
6
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T06:39:52Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
cuongdev/cothu
cuongdev
2024-10-13T06:31:59Z
29
0
diffusers
[ "diffusers", "text-to-image", "stable-diffusion", "license:creativeml-openrail-m", "autotrain_compatible", "endpoints_compatible", "diffusers:StableDiffusionPipeline", "region:us" ]
text-to-image
2024-10-13T06:28:26Z
--- license: creativeml-openrail-m tags: - text-to-image - stable-diffusion --- ### cothu Dreambooth model trained by cuongdev with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) Sample pictures of this concept:
bear7011/llama3.2_1B_red_light
bear7011
2024-10-13T06:03:00Z
17
0
transformers
[ "transformers", "safetensors", "gguf", "llama", "unsloth", "arxiv:1910.09700", "text-generation-inference", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T05:58:51Z
--- library_name: transformers tags: - unsloth --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
josedossantos/urf-txtInteiroTeor-legalbertpt
josedossantos
2024-10-13T05:57:19Z
6
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T05:53:32Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
bear7011/llama3.2_3B_red_light
bear7011
2024-10-13T05:54:03Z
13
0
transformers
[ "transformers", "safetensors", "gguf", "llama", "unsloth", "arxiv:1910.09700", "text-generation-inference", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T05:40:34Z
--- library_name: transformers tags: - unsloth --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
josedossantos/urf-txtInteiroTeor-legalbert
josedossantos
2024-10-13T05:41:39Z
6
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T05:37:51Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
QuantFactory/Reasoning-Llama-3b-v0.1-GGUF
QuantFactory
2024-10-13T05:38:55Z
454
3
transformers
[ "transformers", "gguf", "text-generation-inference", "unsloth", "llama", "trl", "sft", "reasoning", "llama-3", "en", "dataset:KingNish/reasoning-base-20k", "base_model:meta-llama/Llama-3.2-3B-Instruct", "base_model:quantized:meta-llama/Llama-3.2-3B-Instruct", "license:llama3.2", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T05:19:22Z
--- base_model: meta-llama/Llama-3.2-3B-Instruct datasets: - KingNish/reasoning-base-20k language: - en license: llama3.2 tags: - text-generation-inference - transformers - unsloth - llama - trl - sft - reasoning - llama-3 --- [![QuantFactory Banner](https://lh7-rt.googleusercontent.com/docsz/AD_4nXeiuCm7c8lEwEJuRey9kiVZsRn2W-b4pWlu3-X534V3YmVuVc2ZL-NXg2RkzSOOS2JXGHutDuyyNAUtdJI65jGTo8jT9Y99tMi4H4MqL44Uc5QKG77B0d6-JfIkZHFaUA71-RtjyYZWVIhqsNZcx8-OMaA?key=xt3VSDoCbmTY7o-cwwOFwQ)](https://hf.co/QuantFactory) # QuantFactory/Reasoning-Llama-3b-v0.1-GGUF This is quantized version of [KingNish/Reasoning-Llama-3b-v0.1](https://huggingface.co/KingNish/Reasoning-Llama-3b-v0.1) created using llama.cpp # Original Model Card # Model Dexcription It's First iteration of this model. For testing purpose its just trained on 10k rows. It performed very well than expected. It do first reasoning and than generate response on based on it but it do like o1. It do reasoning separately (Just like o1), no tags (like reflection). Below is inference code. ```python from transformers import AutoModelForCausalLM, AutoTokenizer MAX_REASONING_TOKENS = 4096 MAX_RESPONSE_TOKENS = 1024 model_name = "KingNish/Reasoning-Llama-3b-v0.1" model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto") tokenizer = AutoTokenizer.from_pretrained(model_name) prompt = "Which is greater 9.9 or 9.11 ??" messages = [ {"role": "user", "content": prompt} ] # Generate reasoning reasoning_template = tokenizer.apply_chat_template(messages, tokenize=False, add_reasoning_prompt=True) reasoning_inputs = tokenizer(reasoning_template, return_tensors="pt").to(model.device) reasoning_ids = model.generate(**reasoning_inputs, max_new_tokens=MAX_REASONING_TOKENS) reasoning_output = tokenizer.decode(reasoning_ids[0, reasoning_inputs.input_ids.shape[1]:], skip_special_tokens=True) # print("REASONING: " + reasoning_output) # Generate answer messages.append({"role": "reasoning", "content": reasoning_output}) response_template = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) response_inputs = tokenizer(response_template, return_tensors="pt").to(model.device) response_ids = model.generate(**response_inputs, max_new_tokens=MAX_RESPONSE_TOKENS) response_output = tokenizer.decode(response_ids[0, response_inputs.input_ids.shape[1]:], skip_special_tokens=True) print("ANSWER: " + response_output) ``` - **Trained by:** [Nishith Jain](https://huggingface.co/KingNish) - **License:** llama3.2 - **Finetuned from model :** [meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct) - **Dataset used :** [KingNish/reasoning-base-20k](https://huggingface.co/datasets/KingNish/reasoning-base-20k) This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
Rsood/mistral-instruct-v2-phishing-detection-v2
Rsood
2024-10-13T05:27:11Z
78
0
transformers
[ "transformers", "safetensors", "mistral", "text-generation", "conversational", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "4-bit", "bitsandbytes", "region:us" ]
text-generation
2024-10-13T05:25:15Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
josedossantos/urf-summary-legalbertpt
josedossantos
2024-10-13T05:25:53Z
6
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T05:22:05Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
xkang3/distilbert-base-uncased-finetuned-imdb
xkang3
2024-10-13T05:23:39Z
131
0
transformers
[ "transformers", "tensorboard", "safetensors", "distilbert", "fill-mask", "generated_from_trainer", "base_model:distilbert/distilbert-base-uncased", "base_model:finetune:distilbert/distilbert-base-uncased", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]
fill-mask
2024-10-13T04:07:05Z
--- library_name: transformers license: apache-2.0 base_model: distilbert-base-uncased tags: - generated_from_trainer model-index: - name: distilbert-base-uncased-finetuned-imdb results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # distilbert-base-uncased-finetuned-imdb This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.4894 - Model Preparation Time: 0.0032 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 64 - eval_batch_size: 64 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3.0 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | |:-------------:|:-----:|:----:|:---------------:|:----------------------:| | 2.6819 | 1.0 | 157 | 2.4978 | 0.0032 | | 2.5872 | 2.0 | 314 | 2.4488 | 0.0032 | | 2.527 | 3.0 | 471 | 2.4823 | 0.0032 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Tokenizers 0.19.1
josedossantos/urf-summary-legalbertimbau
josedossantos
2024-10-13T05:22:01Z
7
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T05:10:43Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 1024 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
josedossantos/urf-summary-legalbert
josedossantos
2024-10-13T05:10:37Z
6
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "sentence-similarity", "feature-extraction", "dataset_size:10K<n<100K", "loss:ContrastiveLoss", "arxiv:1908.10084", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T05:06:50Z
--- language: [] library_name: sentence-transformers tags: - sentence-transformers - sentence-similarity - feature-extraction - dataset_size:10K<n<100K - loss:ContrastiveLoss widget: - source_sentence: ' O projeto de lei visa alterar o art. 1.788 do Código Civil para normatizar a sucessão de bens e contas digitais do autor da herança de qualidade patrimonial. O projeto pretende tratar do tema relevante e atual de herança digital, que tem sido objeto de diversos casos no Judiciário, e estabelecer medidas para a pacificação de conflitos sociais.' sentences: - ' O Projeto Lei visa corrigir a injustiça que prevê a impenhorabilidade do único imóvel residencial de casal entidade familiar, tornando-o sujeito a penhora para satisfazer dívidas contraídas com o locatário. A Lei estabelece que a impenhorabilidade se aplica somente à dívida civil, comercial e fiscal previdenciária, e não às obrigações decorrentes de contratos de locação. Além disso, a Lei garantiza o direito da família fiador de ser informada sobre a penhora do imóvel e de ter o direito de ser ouvida na decisão da execução.' - ' O projeto de lei nº 2008, do deputado Ratinho Junior, tem como objetivo extinguir a fiança aval prestado por pessoas naturais. A proposta visa eliminar a figura do avalista fiador, que tem sido motivo de ruína para milhões de famílias, e reduzir custos transações comerciais. A justificação da proposta é que a inclusão de pessoas físicas como fiadores é praticamente compulsória e pode levar a prejuízos inocentes. A proposta também busca aumentar a segurança transações e possibilitar a redução de custos para os credores.' - ' O projeto de lei visa reverter as alterações introduzidas pela Reforma Trabalhista de 2017, que afastaram as gestantes e lactantes de quaisquer atividades insalubres, e estabelecer a obrigação de conceder um descanso de 15 minutos entre o fim do horário normal e o início do período de cumprimento de horas extras. Além disso, o projeto busca preservar o direito das trabalhadoras à proteção integral do nascituro e da criança, à saúde e ao meio ambiente de trabalho saudável.' - source_sentence: ' O projeto de lei visa alterar o artigo 45 da Lei nº 8.080, de 19 de setembro de 1990, para permitir o atendimento de serviços saúde para os soldados da Borracha que participaram da Segunda Guerra Mundial, e seus familiares dependentes. O projeto também busca reconhecer o esforço e dedicação dos brasileiros que lutaram na guerra.' sentences: - ' A deputada Perpétua Almeida apresentou um projeto de lei que reconhece os seringueiros soldados borracha que participaram da 2ª Guerra Mundial. O projeto define que esses heróis brasileiros merecem tratamento verdadeiros heróis e devem ser incluídos no Livro Heróis Pátria. O projeto também estabelece que os descendentes dos seringueiros soldados borracha devem ser incluídos no registro de heróis. O projeto foi apresentado na Câmara dos Deputados em 2009.' - ' A Lei nº 11.199/2009, sancionada em 4 de maio de 2009, estabelece a regulamentação da educação superior no Brasil. A lei define a autonomia universitária, o financiamento da educação superior, a assistência estudantil, a extensão e a pesquisa, entre outros aspectos. Além disso, a lei estabelece a criação de fundos para a finança da educação superior e a obrigatoriedade da realização de exames para ingresso nas universidades. A lei também prevê a criação de polos tecnológicos e a ampliação do ensino profissional e técnico.' - ' O projeto de lei nº 4329/2021, apresentado pelo deputado federal Luciano Ducci, visa restringir o uso de cigarros eletrônicos e narguilés em recintos coletivos fechados. A proposta busca inibir o consumo desses produtos, que podem ser prejudiciais à saúde, especialmente para o público jovem. A justificativa é que o Brasil tem sido um dos países mais bem-sucedidos na redução do tabagismo, mas a indústria do tabaco tem encontrado artifícios para driblar as restrições. O projeto busca submeter os cigarros eletrônicos e narguilés às mesmas restrições aplicadas aos demais produtos fumígenos.' - source_sentence: ' O projeto de lei visa permitir que os contribuintes deduzam das receitas tributáveis as despesas realizadas com matrículas e mensalidades de cursos de idiomas estrangeiros, com o objetivo de incentivar a educação e o desenvolvimento profissional. A proposta foi apresentada pelo deputado Fernando Coruja e está sendo discutida no Congresso Nacional.' sentences: - ' O projeto de lei nº 2016, de 2016, proposto pelo deputado Aureo, permite a dedução de pagamentos relativos a despesas com cursos de idiomas estrangeiros para apuração do imposto sobre a renda das pessoas físicas. A justificativa para a proposta é que a rede pública de ensino não é capaz de dotar os brasileiros das habilidades necessárias para dominar um idioma estrangeiro, e que a dedução deveria incentivar o custeio dos cursos realizados em escolas privadas. O projeto prevê a dedução de despesas com cursos de idiomas estrangeiros até o limite anual individual de NR$ 2.000,00.' - ' O projeto de lei visa regular o cancelamento de bilhetes aéreos de ida e volta em caso de não comparecimento do passageiro para embarque, garantindo a validade e exequibilidade do trecho de retorno. A proposta estabelece uma limitação de 50% do valor adimplido pelos passageiros para taxas de cancelamento e remarcação, e prevê punições para descumprimento da lei. A proposta busca proteger os consumidores e garantir a equidade nas relações comerciais entre os prestadores de serviços de transporte aéreo e os passageiros.' - ' A Lei nº 10.452/2006, aprovada pelo Congresso Nacional em 2006, alterou o art. 193 da Consolidação das Leis do Trabalho (CLT) para considerar as atividades de carteiros como operações perigosas, regulamentadas pelo Ministério do Trabalho e Emprego. Essa alteração entrou em vigor em julho de 2006, após a publicação no Senado Federal. A lei estabelece que as atividades de carteiros exercidas em condições de risco à integridade física do trabalhador, decorrência da circulação de vias públicas perigosas e entrega de correspondência em comum, são consideradas operações perigosas.' - source_sentence: ' O projeto de lei visa extinguir as enfiteuses existentes e modernizar a legislação civil brasileira, permitindo a transferência do domínio útil e a cessão do senhorio, e estabelecendo novos parâmetros jurídicos para o instituto da enfiteuse. A lei também regulamenta a terrenos marinha e estabelece novos critérios para a definição desses terrenos.' sentences: - ' O projeto de lei nº 2010, do deputado Ricardo Tripoli, objetiva extinção da cobrança de laudêmio em transmissões aforadas, com base na justificativa de que o instituto da enfiteuse não é mais necessário e que a cobrança do laudêmio é uma onera que não corresponde à realidade, pois o proprietário da propriedade already possui o domínio útil e direto. Além disso, o projeto também propõe a extinção da cobrança de laudêmio em caso de doação, desapropriação e herança, com o objetivo de aperfeiçoar o ordenamento jurídico em relação a essas transações.' - ' O projeto de lei proposto por Laerte Bessa, PMDB-DF, visa regulamentar a jornada de trabalho de profissionais de saúde, estabelecendo um limite de 80 horas semanais para esses profissionários. A proposta é baseada na Constituição e na necessidade de garantir a qualidade do trabalho e a dignidade dos profissionais de saúde, que atualmente enfrentam demandas exageradas e desumanas. A medida busca evitar a contratação de servidores empregados em horários extras e rotineiros, garantindo o respeito ao limite de jornada semanal.' - ' O projeto de lei apresentado por Raul Henry em 2007 visa estabelecer um prazo para a proibição da produção, comercialização e utilização de pilhas recarregáveis, com o objetivo de evitar a contaminação causada por esses resíduos perigosos. A proposta estabelece penalidades para quem não cumpriser as normas ambientais e propõe educativas medidas para o descarte correto dos materiais tóxicos. A iniciativa busca melhorar o equilíbrio ecológico do planeta e reduzir o prejuízo ecológico causado pela utilização de pilhas recarregáveis.' - source_sentence: ' O projeto de lei visa alterar o art. 20 da Lei nº 5.474/68, permitindo que as empresas locadoras de bens móveis emitam faturas duplicatas para representar créditos oriundos de contratos de locação de bens móveis. A proposta tem como objetivo regularizar a prática comercial usual e evitar futuros questionamentos no Poder Judiciário.' sentences: - ' A proposta de lei visa permitir a emissão de fatura e duplicata a partir da locação de bens móveis, além de estabelecer que a locação de bens móveis não se equipara à prestação de serviços. A justificativa é que a emissão de duplicata é necessária para a adequada administração financeira das empresas e para viabilizar o desconto do título de crédito por meio do sistema bancário. Além disso, a proposta de lei não alterará o art. 26 da Lei nº 5.474, de 1968, que estabeleceu a nova redação do Código Penal.' - ' O Projeto de Lei apresentado pelo Deputado Cabo Júlio estabelece normas para o atendimento de vítimas de violência sexual em hospitais públicos e privados, obrigando a notificação compulsória e o registro de dados identificados das vítimas. A proposta busca garantir o direito à saúde e à dignidade das vítimas, além de combater a impunidade e a omissão de responsabilidades por parte das autoridades. A medida também pretende incentivar a prevenção da violência sexual e a identificação e condicionamento de agressores.' - ' O projeto de lei proíbe a venda de refrigerantes a menores de quatorze anos, por serem prejudiciais à saúde. A justificativa é que os refrigerantes não têm valor nutricional, são fontes de açúcar e podem causar obesidade, cáries dentárias e irritação na mucosa digestiva. Além disso, as versões dietéticas têm edulcorantes de segurança duvidosa. O projeto visa proteger as crianças de tais efeitos prejudiciais e criar novas gerações mais saudáveis.' pipeline_tag: sentence-similarity --- # SentenceTransformer This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer <!-- - **Base model:** [Unknown](https://huggingface.co/unknown) --> - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 768 tokens - **Similarity Function:** Cosine Similarity <!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown --> ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("josedossantos/urf-summary-legalbert") # Run inference sentences = [ ' O projeto de lei visa alterar o art. 20 da Lei nº 5.474/68, permitindo que as empresas locadoras de bens móveis emitam faturas duplicatas para representar créditos oriundos de contratos de locação de bens móveis. A proposta tem como objetivo regularizar a prática comercial usual e evitar futuros questionamentos no Poder Judiciário.', ' A proposta de lei visa permitir a emissão de fatura e duplicata a partir da locação de bens móveis, além de estabelecer que a locação de bens móveis não se equipara à prestação de serviços. A justificativa é que a emissão de duplicata é necessária para a adequada administração financeira das empresas e para viabilizar o desconto do título de crédito por meio do sistema bancário. Além disso, a proposta de lei não alterará o art. 26 da Lei nº 5.474, de 1968, que estabeleceu a nova redação do Código Penal.', ' O Projeto de Lei apresentado pelo Deputado Cabo Júlio estabelece normas para o atendimento de vítimas de violência sexual em hospitais públicos e privados, obrigando a notificação compulsória e o registro de dados identificados das vítimas. A proposta busca garantir o direito à saúde e à dignidade das vítimas, além de combater a impunidade e a omissão de responsabilidades por parte das autoridades. A medida também pretende incentivar a prevenção da violência sexual e a identificação e condicionamento de agressores.', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 768] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` <!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* --> ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 10,962 training samples * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code> * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | label | |:--------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 49 tokens</li><li>mean: 141.66 tokens</li><li>max: 344 tokens</li></ul> | <ul><li>min: 36 tokens</li><li>mean: 140.06 tokens</li><li>max: 355 tokens</li></ul> | <ul><li>0: ~49.50%</li><li>1: ~50.50%</li></ul> | * Samples: | sentence_0 | sentence_1 | label | |:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------| | <code> A Lei nº 152, de 2022, alterou a Lei nº 12.187, de 2009, estabelecendo novas metas de compromisso nacional voluntário para reduzir emissões de gases de efeito estufa até 2025 e 2030, em relação a 2005. Essas metas serão regulamentadas e depositadas junto à Convenção-Quadro das Nações Unidas sobre Mudança do Clima. A Lei também destaca a importância da eliminação do desmatamento ilegal e da promoção da agropecuária sustentável.</code> | <code> O projeto de lei estabelece preferência no desempate de processo licitatório para empresas que comprovem mitigação à mudança do clima. A proposta inclui a inserção de inciso VI ao § 2º do art. 3º da Lei nº 8.666, de 21 de junho de 1993, que regulamenta o art. 37, inciso XXI, da Constituição Federal, e dá outras providências. A mitigação ambiental é entendida como mudanças e substituições tecnológicas que reduzam o uso de recursos e as emissões por unidade de produção, bem como a implementação de medidas que reduzam as emissões de gases de efeito estufa e aumentem os sumidouros. A proposta busca contribuir para o combate ao aquecimento global e à poluição ambiental, sem resultar em prejuízos ao Poder Público ou à sociedade.</code> | <code>0</code> | | <code> O projeto de lei visa estabelecer normas para a inserção de contratos honorários em causas criminais, obrigando os advogados a comprovar a origem lícita dos recursos utilizados e a identificar o caso pessoa responsável. A lei busca prevenir o uso de dinheiro sujo em pagamentos honorários e contribuição para a administração da justiça.</code> | <code> O projeto de lei visa obrigar os réus de crimes hediondos e organizações criminosas a demonstrar a origem lícita dos recursos financeiros usados para pagar honorários de advogados. A proposta busca salvaguardar a profissão honesta e transparente, evitando que os advogados sejam usados para lavar dinheiro de atividades criminosas. O projeto também prevê a possibilidade de um juiz verificar indícios de fraude e comunicação com o Conselho Controle Atividades Financeiras.</code> | <code>1</code> | | <code> O presente projeto de lei institui o Estatuto do Carbono Verde, regulamentando o mercado de carbono no âmbito dos povos tradicionais e do agronegócio, com foco em reduzir as emissões de gases de efeito estufa e promover o desenvolvimento sustentável. A proposta traz novas medidas, como a criação da Secretaria Especial do Carbono Verde, o Programa Nacional Obrigatória de Compensão de Emissões de GEE e o Plano Nacional Agrícola de Redução das Emissões de GEE. Além disso, a proposta estabelece a comercialização de créditos de carbono nos mercados obrigatórios, permitindo a inclusão de povos tradicionais e comunidades tradicionais na comercialização dos créditos.<br> O presente projeto de lei é uma iniciativa inovante para o desenvolvimento sustentável do país, garantindo o direito dos povos tradicionais e comunidades tradicionais a participar no mercado obrigatória de carbono verde mercado. Apresenta a proposto de lei nº 121218/20009, de 20009/201</code> | <code> A proposta de lei visa estabelecer incentivos fiscais para pessoas físicas e jurídicas que investem em projetos de desenvolvimento limpo, como o mercado de reduções certificadas de emissões (RCEs). O projeto prevê a criação de fundos de investimento para financiar esses projetos e a regulamentação da Comissão de Valores Mobiliários. A lei também estabelece que as quotas do Fundo de Investimento em Projetos de Desenvolvimento Limpo (FIMDL) serão isentas de imposto de renda e outras taxas. Além disso, a lei prevê a compensação de prejuízos para investidores residentes e domiciliados no exterior que venham a sofrer prejuízos decorrentes da alienação de quotas do FIMDL.</code> | <code>0</code> | * Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters: ```json { "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", "margin": 0.5, "size_average": true } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters <details><summary>Click to expand</summary> - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin </details> ### Training Logs | Epoch | Step | Training Loss | |:------:|:----:|:-------------:| | 0.0912 | 500 | 0.0399 | | 0.1824 | 1000 | 0.0263 | | 0.2737 | 1500 | 0.0236 | | 0.3649 | 2000 | 0.0217 | | 0.4561 | 2500 | 0.0189 | | 0.5473 | 3000 | 0.0179 | | 0.6386 | 3500 | 0.0155 | | 0.7298 | 4000 | 0.0137 | | 0.8210 | 4500 | 0.011 | | 0.9122 | 5000 | 0.0104 | ### Framework Versions - Python: 3.10.14 - Sentence Transformers: 3.0.0 - Transformers: 4.39.3 - PyTorch: 2.2.0 - Accelerate: 0.30.1 - Datasets: 2.14.4 - Tokenizers: 0.15.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### ContrastiveLoss ```bibtex @inproceedings{hadsell2006dimensionality, author={Hadsell, R. and Chopra, S. and LeCun, Y.}, booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, title={Dimensionality Reduction by Learning an Invariant Mapping}, year={2006}, volume={2}, number={}, pages={1735-1742}, doi={10.1109/CVPR.2006.100} } ``` <!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->
vkunchur19/whisper-medium-konnakol-rests
vkunchur19
2024-10-13T05:08:18Z
6
0
transformers
[ "transformers", "tensorboard", "safetensors", "whisper", "automatic-speech-recognition", "generated_from_trainer", "dataset:audiofolder", "base_model:openai/whisper-medium", "base_model:finetune:openai/whisper-medium", "license:apache-2.0", "model-index", "endpoints_compatible", "region:us" ]
automatic-speech-recognition
2024-10-13T01:10:42Z
--- library_name: transformers license: apache-2.0 base_model: openai/whisper-medium tags: - generated_from_trainer datasets: - audiofolder metrics: - wer model-index: - name: whisper-medium-konnakol-rests results: - task: name: Automatic Speech Recognition type: automatic-speech-recognition dataset: name: audiofolder type: audiofolder config: default split: test args: default metrics: - name: Wer type: wer value: 25.995316159250585 --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # whisper-medium-konnakol-rests This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the audiofolder dataset. It achieves the following results on the evaluation set: - Loss: 0.0952 - Wer: 25.9953 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 2 - eval_batch_size: 1 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 16 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 100 - training_steps: 1000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:--------:|:----:|:---------------:|:-------:| | 0.6638 | 13.3333 | 50 | 0.0648 | 50.5855 | | 0.0132 | 26.6667 | 100 | 0.0678 | 40.7494 | | 0.0068 | 40.0 | 150 | 0.0794 | 30.6792 | | 0.0033 | 53.3333 | 200 | 0.0909 | 28.3372 | | 0.0008 | 66.6667 | 250 | 0.0821 | 25.9953 | | 0.0021 | 80.0 | 300 | 0.0725 | 29.5082 | | 0.0002 | 93.3333 | 350 | 0.0933 | 26.2295 | | 0.0 | 106.6667 | 400 | 0.0932 | 26.2295 | | 0.0 | 120.0 | 450 | 0.0937 | 26.2295 | | 0.0 | 133.3333 | 500 | 0.0940 | 26.2295 | | 0.0 | 146.6667 | 550 | 0.0942 | 26.2295 | | 0.0 | 160.0 | 600 | 0.0944 | 25.9953 | | 0.0 | 173.3333 | 650 | 0.0947 | 25.9953 | | 0.0 | 186.6667 | 700 | 0.0946 | 25.9953 | | 0.0 | 200.0 | 750 | 0.0947 | 25.9953 | | 0.0 | 213.3333 | 800 | 0.0948 | 25.9953 | | 0.0 | 226.6667 | 850 | 0.0950 | 25.9953 | | 0.0 | 240.0 | 900 | 0.0950 | 25.9953 | | 0.0 | 253.3333 | 950 | 0.0952 | 25.9953 | | 0.0 | 266.6667 | 1000 | 0.0952 | 25.9953 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.19.1
Ayon128/t5_base-2
Ayon128
2024-10-13T05:08:07Z
114
0
transformers
[ "transformers", "safetensors", "t5", "text2text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text2text-generation
2024-10-13T05:07:29Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
josedossantos/urf-summary-bertimbau
josedossantos
2024-10-13T05:06:46Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "sentence-similarity", "feature-extraction", "dataset_size:10K<n<100K", "loss:ContrastiveLoss", "arxiv:1908.10084", "base_model:neuralmind/bert-large-portuguese-cased", "base_model:finetune:neuralmind/bert-large-portuguese-cased", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T04:55:43Z
--- language: [] library_name: sentence-transformers tags: - sentence-transformers - sentence-similarity - feature-extraction - dataset_size:10K<n<100K - loss:ContrastiveLoss base_model: neuralmind/bert-large-portuguese-cased widget: - source_sentence: ' O projeto de lei visa alterar o art. 1.788 do Código Civil para normatizar a sucessão de bens e contas digitais do autor da herança de qualidade patrimonial. O projeto pretende tratar do tema relevante e atual de herança digital, que tem sido objeto de diversos casos no Judiciário, e estabelecer medidas para a pacificação de conflitos sociais.' sentences: - ' O projeto de lei visa alterar o art. 1.788 do Código Civil, estabelecendo que os herdeiros terão direito acesso e controle sobre contas digitais e arquivos de serviços online de falecidos, sem precisar de autorização judicial. A medida busca regularizar e uniformizar o tratamento das sucessões bens contas digitais, evitando conflitos sociais e garantindo o direito dos herdeiros.' - ' O projeto de lei visa alterar a redação do Art. 43 da Lei nº 8.078/90, permitindo a inclusão de cadastros negativos de consumidores em bancos de dados, contestando débitos líquidos certos sendo contestados judicialmente. Essa alteração busca adequar as disposições legais às novas necessidades do mercado e evitar a exclusão de consumidores dos bancos de dados. Ao permitir a inclusão de cadastros negativos, o projeto busca evitar a pressão de fornecedores para que os consumidores reconheçam débitos que não são verdadeiros.' - ' O projeto de lei proíbe as centrais de telemarketing de realizar chamadas automatizadas sem a presença de um operador natural. As chamadas devem ter a opção de contato com um atendente natural no primeiro menu eletrônico. A fiscalização será realizada pelo Sistema Nacional de Defesa do Consumidor do Ministério da Justiça. A aplicação da multa de até R$ 2.000.000,00 pode ser aplicada em caso de descumprimento. O projeto visa proteger os cidadãos brasileiros de chamadas agressivas e invasivas.' - source_sentence: ' O projeto de lei proíbe a realização de chamadas telefônicas anônimas, obrigando a identificação do remetente e do destinatário. A justificativa é a necessidade de combater crimes e violações de direitos garantidos, que ocorrem por meio do uso inadequado da telefonia. A proibição será aplicada a partir da data de publicação da lei.' sentences: - ' O projeto de lei visa obrigar as operadoras de telefonia fixa e celular a informarem o código de acesso do usuário chamador em ligações telefônicas, sem ônus para o usuário receptor. A medida busca reforçar a segurança pública e facilitar as investigações e a prisão de criminosos que se escondem atrás da situação anônima da chamada não identificada. O projeto foi apresentado pelo deputado federal Alberto Fraga e justifica-se com a necessidade de combater a criminalidade que utiliza as telecomunicações para perpetrar crimes.' - ' O projeto de lei nº 2007 do sr deputado Deley estabelece concessão de visto temporário para estrangeiros que vierem trabalhar temporariamente em eventos esportivos internacionais no Brasil, como a Copa do Mundo de 2014 e as Olimpíadas. A lei também estabelece facilidades para a concessão de visto e registro de entrada no país para essas pessoas. A justificativa para o projeto é a importância dos eventos esportivos brasileiros e a necessidade de garantir a presença de estrangeiros para a realização desses eventos.' - ' O projeto de lei visa alterar a redação do Art. 43 da Lei nº 8.078/90, permitindo a inclusão de cadastros negativos de consumidores em bancos de dados, contestando débitos líquidos certos sendo contestados judicialmente. Essa alteração busca adequar as disposições legais às novas necessidades do mercado e evitar a exclusão de consumidores dos bancos de dados. Ao permitir a inclusão de cadastros negativos, o projeto busca evitar a pressão de fornecedores para que os consumidores reconheçam débitos que não são verdadeiros.' - source_sentence: ' O projeto de lei visa permitir que os contribuintes deduzam das receitas tributáveis as despesas realizadas com matrículas e mensalidades de cursos de idiomas estrangeiros, com o objetivo de incentivar a educação e o desenvolvimento profissional. A proposta foi apresentada pelo deputado Fernando Coruja e está sendo discutida no Congresso Nacional.' sentences: - ' O projeto de lei nº 2016, de 2016, proposto pelo deputado Aureo, permite a dedução de pagamentos relativos a despesas com cursos de idiomas estrangeiros para apuração do imposto sobre a renda das pessoas físicas. A justificativa para a proposta é que a rede pública de ensino não é capaz de dotar os brasileiros das habilidades necessárias para dominar um idioma estrangeiro, e que a dedução deveria incentivar o custeio dos cursos realizados em escolas privadas. O projeto prevê a dedução de despesas com cursos de idiomas estrangeiros até o limite anual individual de NR$ 2.000,00.' - ' O presente Projeto de Lei visa reformular o Ensino Médio, com o objetivo de torná-lo mais atraente e de qualidade, oferecendo opções formativas para os alunos, com ênfase em línguas, matemáticas, ciências e humanidades, além de habilidades técnicas e sociais. Além disso, estabelece a jornada escolar em tempo integral, permitindo a ampliação das vagas de ensino médio e a redução da taxa de abandono. O projeto também prevê a criação de programas de formação para os professores e a revisão dos conteúdos curriculares.' - ' O projeto de lei dispõe que as operadoras telefonia móvel devem manter um cadastro atualizado com informações dos proprietários dos aparelhos celulares, e que os usuários devem fornecer cópias simples das informações de identificação para a operadora. Além disso, o projeto estabelece penalidades para a utilização indevida dos aparelhos celulares, e prevê a bloqueio do IMEI do aparelho em caso de perda, furto ou roubo. O objetivo do projeto é proteger o patrimônio dos cidadãos e garantir a segurança pública.' - source_sentence: ' O projeto de lei do deputado Washington Reis dispõe isenção ao pagamento de pedágio para motocicletas similares em rodovias federais. A justificativa para essa proposição é que esses veículos causam menos danos às estradas e ao meio ambiente, além de representarem uma economia de gasolina e uma diminuição no volume do tráfego.' sentences: - ' O projeto de lei proposto por Deputado Gutemberg Reis dispõe sobre a isenção de pagamento de pedágio para motocicletas e similares em rodovias federais. A justificativa para essa proposta é que esses veículos, devido às suas características, não causam danos às estradas e ao meio-ambiente, e que eles representam economia de gasolina, diminuição do volume de tráfego e diminuição dos índices de poluição. O projeto de lei busca contemplar esses veículos com uma isenção de pagamento de pedágio em rodovias federais.' - ' O projeto de lei visa alterar o Código Eleitoral para instituir o sistema majoritário uninominal nas eleições para vereadores em municípios com mais de 200.000 eleitores, permitindo que os eleitores tenham certeza de que o seu voto tem efeito direto na escolha do vereador e reduzindo custos de financiamento das campanhas. Além disso, a proposta prevê a extinção do horário eleitoral gratuito no rádio e na televisão para os candidatos a vereador, considerando que o sistema distrital será mais eficaz e menos dispendioso.' - ' O projeto de lei proíbe a realização de ligações telefônicas não identificáveis, com o objetivo de inibir a proliferação de ligações confidenciais com fins de telemarketing e cobrança, e também para reduzir os índices de criminalidade no país. O projeto mantém o dispositivo da Lei Geral de Telecomunicações que assegura ao usuário o direito à não divulgação do seu número telefônico, mas apenas em listas de assinantes. O projeto estabelece um prazo de noventa dias para que a norma proposta passe a vigorar, permitindo que as operadoras adaptem suas redes para adequação ao disposto na proposição.' - source_sentence: ' O projeto de lei visa alterar o artigo 45 da Lei nº 8.080, de 19 de setembro de 1990, para permitir o atendimento de serviços saúde para os soldados da Borracha que participaram da Segunda Guerra Mundial, e seus familiares dependentes. O projeto também busca reconhecer o esforço e dedicação dos brasileiros que lutaram na guerra.' sentences: - ' O projeto de lei apresentado pelo deputado Mauro Nazif em 2007 visa sanar uma falha legislativa na Lei nº 7.986/1989, que omite a gratificação natalina para beneficiários de pensões vitalícias. O projeto propõe a alteração da Lei nº 7.986/1989 para garantir o recebimento da gratificação natalina para os beneficiários de pensões vitalícias, que atualmente não têm direito a essa benefício. Além disso, o projeto também busca regular a percepção da gratificação natalina para os aposentados e anistiaados.' - ' A proposta de emenda constitucional visa criar as polícias penais federal, estaduais e distrital, e estabelecer as suas atribuições e subordinação. O quadro de servidores das polícias penais será preenchido por meio de concurso público ou da transformação dos cargos isolados ou dos cargos de carreira dos atuais agentes penitenciários ou equivalentes. A emenda também estabelece a subordinação das polícias militares e dos corpos de bombeiros militares aos Governadores dos Estados, do Distrito Federal e dos Territórios.' - ' A proposta de lei visa acrescentar as metas brasileiras de redução de emissões para os períodos pós-2020, e dá outras providências. O Congresso Nacional decreta que a Política Nacional sobre Mudança do Clima será instituída, e as metas brasileiras de redução de emissões propostas para os períodos de 2020 a 2025 e de 2025 a 2030. Além disso, a Lei acrescenta o art. 12-A, que estabelece que as propostas brasileiras sobre prevenção, mitigação e adaptação às mudanças do clima devem ser precedidas de consulta pública e divulgadas em todo o território nacional.' pipeline_tag: sentence-similarity --- # SentenceTransformer based on neuralmind/bert-large-portuguese-cased This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased) <!-- at revision aa302f6ea73b759f7df9cad58bd272127b67ec28 --> - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 1024 tokens - **Similarity Function:** Cosine Similarity <!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown --> ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("josedossantos/urf-summary-bertimbau") # Run inference sentences = [ ' O projeto de lei visa alterar o artigo 45 da Lei nº 8.080, de 19 de setembro de 1990, para permitir o atendimento de serviços saúde para os soldados da Borracha que participaram da Segunda Guerra Mundial, e seus familiares dependentes. O projeto também busca reconhecer o esforço e dedicação dos brasileiros que lutaram na guerra.', ' O projeto de lei apresentado pelo deputado Mauro Nazif em 2007 visa sanar uma falha legislativa na Lei nº 7.986/1989, que omite a gratificação natalina para beneficiários de pensões vitalícias. O projeto propõe a alteração da Lei nº 7.986/1989 para garantir o recebimento da gratificação natalina para os beneficiários de pensões vitalícias, que atualmente não têm direito a essa benefício. Além disso, o projeto também busca regular a percepção da gratificação natalina para os aposentados e anistiaados.', ' A proposta de emenda constitucional visa criar as polícias penais federal, estaduais e distrital, e estabelecer as suas atribuições e subordinação. O quadro de servidores das polícias penais será preenchido por meio de concurso público ou da transformação dos cargos isolados ou dos cargos de carreira dos atuais agentes penitenciários ou equivalentes. A emenda também estabelece a subordinação das polícias militares e dos corpos de bombeiros militares aos Governadores dos Estados, do Distrito Federal e dos Territórios.', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 1024] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` <!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* --> ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 10,962 training samples * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code> * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | label | |:--------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 44 tokens</li><li>mean: 142.17 tokens</li><li>max: 344 tokens</li></ul> | <ul><li>min: 36 tokens</li><li>mean: 137.21 tokens</li><li>max: 355 tokens</li></ul> | <ul><li>0: ~49.00%</li><li>1: ~51.00%</li></ul> | * Samples: | sentence_0 | sentence_1 | label | |:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------| | <code> O Projeto de Lei do Deputado Federal Celso Sabino (PSDB/PA) objetiva coibir o comportamento de furar a fila na ordem de prioridades para vacinação contra COVID-19 e outras enfermidades, estabelecidas por determinadas campanhas de imunização. O projeto propõe a alteração do Decreto-Lei nº 2.848, de 7 de dezembro de 1940, Código Penal, para tornar crime a violação de prioridade para vacinação. O projeto também prevê a criação de um tipo penal específico para a conduta ignominiosa de furar a fila da ordem de prioridade para vacinação, com pena de detenção de um mês a um ano e multa. Além disso, o projeto estabelece que a pena será aumentada se o agente for um funcionário público e cometer o crime, prevalecendo-se do cargo. O projeto foi apresentado na Câmara dos Deputados em 29 de janeiro de 2021 e conta com o apoio de outros deputados.</code> | <code> O projeto de lei nº 1519/2021, apresentado pelo deputado Paulo Bengtson, prioriza o atendimento dos portadores de diabetes nos laboratórios das redes públicas e privadas de saúde durante a realização de exames que necessitem de jejum total. A lei estabelece que os portadores de diabetes têm prioridade na fila de atendimento, juntamente com as gestantes, idosos e pessoas com deficiência. O projeto também determina que os pacientes devem apresentar documentos médicos que comprovem a patologia para ter direito ao tratamento prioritário. A justificativa para o projeto é evitar mal-estar e complicações para os diabéticos, que podem desencadear malefícios à saúde se ficarem muito tempo sem se alimentar. A lei entrará em vigor na data de sua publicação.</code> | <code>0</code> | | <code> O projeto de Lei Institui a obrigatoriedade de estabelecimentos públicos e privados voltados ao ensino ou recreação infantil e fundamental a capacitarem seu corpo docente e funcional em noções básicas de primeiros socorros, para garantir a integridade física e psicológica de seus tutelados e alunos, evitando sequelas fisiológicas e anatômicas irremediáveis. O curso será ministrado por entidades municipais ou estaduais, especializadas em práticas de auxílio imediato e emergencial à população, e as despesas serão orçamentárias. O não cumprimento dos dispositivos da Lei, implicará às instituições de ensino inadimplentes.</code> | <code> O Projeto de Lei nº 2015/6891, de 2015, propõe a adição de um artigo ao Decreto nº 9.394/1996, que estabelece as diretrizes e bases da educação nacional, para instituição da obrigatoriedade do ensino de primeiros socorros nos cursos e programas da educação superior. O ensino deverá ter uma duração mínima de 100 horas e ser ministrado por profissionais com formação na área da saúde. A aprovação na disciplina de noções de primeiros socorros será condição para a obtenção do diploma ou certificado de conclusão dos cursos e programas da educação superior. O objetivo do Projeto é reduzir o número de óbitos e lesões graves causadas por acidentes, ao ensejar as noções básicas de primeiros socorros para os estudantes de educação superior.</code> | <code>1</code> | | <code> O Projeto de Lei nº 2.004, de 2004, do Deputado Paulo Delgado, estabelece a tributação de lucros extraordinários obtidos por instituições financeiras, com o objetivo de garantir a justiça fiscal e obter recursos para o pagamento de aposentadorias e programas de estabilização. A proposta estabelece uma adicional de 18% sobre o lucro líquido apurado, com base no patrimônio líquido médio das instituições financeiras. A medida busca evitar que as instituições financeiras auferam lucros extraordinários à custa dos programas de estabilização e da renda dos cidadãos. A justificativa da proposta é que os lucros extraordinários obtidos pelos instituições financeiras são fruto de políticas governamentais de estabilização e fiscalização, e que esses lucros devem ser revertidos para a sociedade, através da contribuição social.</code> | <code> O projeto de lei nº 2005, do deputado Carlos Nader, dispõe obrigatoriedade de contratação de seguro para estacionamentos privados, com penalidades para não cumprimento. O seguro deve garantir indenização por danos a veículos e pessoas, e o comprovante de estacionamento deve indicar nome da seguradora e apólice. A lei também estabelece sanções para não cumprimento e justifica a necessidade de proteger direitos dos consumidores.</code> | <code>0</code> | * Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters: ```json { "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", "margin": 0.5, "size_average": true } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters <details><summary>Click to expand</summary> - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin </details> ### Training Logs | Epoch | Step | Training Loss | |:------:|:----:|:-------------:| | 0.0912 | 500 | 0.0372 | | 0.1824 | 1000 | 0.0216 | | 0.2737 | 1500 | 0.019 | | 0.3649 | 2000 | 0.0163 | | 0.4561 | 2500 | 0.0143 | | 0.5473 | 3000 | 0.0122 | | 0.6386 | 3500 | 0.0111 | | 0.7298 | 4000 | 0.0092 | | 0.8210 | 4500 | 0.0086 | | 0.9122 | 5000 | 0.0072 | ### Framework Versions - Python: 3.10.14 - Sentence Transformers: 3.0.0 - Transformers: 4.39.3 - PyTorch: 2.2.0 - Accelerate: 0.30.1 - Datasets: 2.14.4 - Tokenizers: 0.15.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### ContrastiveLoss ```bibtex @inproceedings{hadsell2006dimensionality, author={Hadsell, R. and Chopra, S. and LeCun, Y.}, booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, title={Dimensionality Reduction by Learning an Invariant Mapping}, year={2006}, volume={2}, number={}, pages={1735-1742}, doi={10.1109/CVPR.2006.100} } ``` <!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->
Ayon128/t5_base-0
Ayon128
2024-10-13T05:04:44Z
116
0
transformers
[ "transformers", "safetensors", "t5", "text2text-generation", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text2text-generation
2024-10-13T05:04:09Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
ammonbro/bart_updown_sp
ammonbro
2024-10-13T05:04:39Z
169
0
transformers
[ "transformers", "safetensors", "bart", "text-generation", "generated_from_trainer", "base_model:facebook/bart-base", "base_model:finetune:facebook/bart-base", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T04:56:04Z
--- library_name: transformers license: apache-2.0 base_model: facebook/bart-base tags: - generated_from_trainer model-index: - name: bart_updown_sp results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # bart_updown_sp This model is a fine-tuned version of [facebook/bart-base](https://huggingface.co/facebook/bart-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.0 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:-----:|:---------------:| | 0.0 | 1.0 | 1429 | 0.0000 | | 0.0 | 2.0 | 2858 | 0.0000 | | 0.0 | 3.0 | 4287 | 0.0000 | | 0.0 | 4.0 | 5716 | 0.0000 | | 0.0 | 5.0 | 7145 | 0.0 | | 0.0 | 6.0 | 8574 | 0.0 | | 0.0 | 7.0 | 10003 | 0.0 | | 0.0 | 8.0 | 11432 | 0.0 | | 0.0 | 9.0 | 12861 | 0.0 | | 0.0 | 10.0 | 14290 | 0.0 | ### Framework versions - Transformers 4.45.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.20.1
Statuo/ArliAI-MN-12b-v1.2-EXL2-4bpw
Statuo
2024-10-13T05:01:24Z
5
0
null
[ "safetensors", "mistral", "license:apache-2.0", "4-bit", "exl2", "region:us" ]
null
2024-10-13T04:48:14Z
--- license: apache-2.0 --- # ArliAI-RPMax-12B-v1.2 ===================================== Saw ArliAI came out with a new version of the MN model and since I quanted it into EXL2 before figured I'd do it again. Though tbh I prefer Stardust v2, it's possible that this new update fixes the issues I had with it previously. Who knows? [This is the EXL2 4bpw version of this model. Find the original model here.](https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2) <br> [Find the 8bpw version here.](https://huggingface.co/Statuo/ArliAI-MN-12b-v1.2-EXL2-8bpw) <br> [Find the 6bpw version here.](https://huggingface.co/Statuo/ArliAI-MN-12b-v1.2-EXL2-6bpw) <br> ## UPDATE: For those getting gibberish results, it was merged wrongly to base after LORA training. Reuploaded all the files so it should work properly now. ## RPMax Series Overview | [2B](https://huggingface.co/ArliAI/Gemma-2-2B-ArliAI-RPMax-v1.1) | [3.8B](https://huggingface.co/ArliAI/Phi-3.5-mini-3.8B-ArliAI-RPMax-v1.1) | [8B](https://huggingface.co/ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.2) | [9B](https://huggingface.co/ArliAI/Gemma-2-9B-ArliAI-RPMax-v1.1) | [12B](https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2) | [20B](https://huggingface.co/ArliAI/InternLM2_5-20B-ArliAI-RPMax-v1.1) | [22B](https://huggingface.co/ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1) | [70B](https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1) | RPMax is a series of models that are trained on a diverse set of curated creative writing and RP datasets with a focus on variety and deduplication. This model is designed to be highly creative and non-repetitive by making sure no two entries in the dataset have repeated characters or situations, which makes sure the model does not latch on to a certain personality and be capable of understanding and acting appropriately to any characters or situations. Early tests by users mentioned that these models does not feel like any other RP models, having a different style and generally doesn't feel in-bred. You can access the model at https://arliai.com and ask questions at https://www.reddit.com/r/ArliAI/ We also have a models ranking page at https://www.arliai.com/models-ranking Ask questions in our new Discord Server! https://discord.com/invite/t75KbPgwhk ## Model Description ArliAI-RPMax-12B-v1.2 is a variant based on Mistral Nemo 12B Instruct 2407. This is arguably the most successful RPMax model due to how Mistral is already very uncensored in the first place. v1.2 update completely removes non-creative/RP examples in the dataset and is also an incremental improvement of the RPMax dataset which dedups the dataset even more and better filtering to cutout irrelevant description text that came from card sharing sites. ### Specs * **Context Length**: 128K * **Parameters**: 12B ### Training Details * **Sequence Length**: 8192 * **Training Duration**: Approximately 2 days on 2x3090Ti * **Epochs**: 1 epoch training for minimized repetition sickness * **LORA**: 64-rank 128-alpha, resulting in ~2% trainable weights * **Learning Rate**: 0.00001 * **Gradient accumulation**: Very low 32 for better learning. ## Quantization The model is available in quantized formats: * **FP16**: https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2 * **GGUF**: https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2-GGUF ## Suggested Prompt Format Mistral Instruct Prompt Format
Statuo/ArliAI-MN-12b-v1.2-EXL2-8bpw
Statuo
2024-10-13T04:58:47Z
12
0
null
[ "safetensors", "mistral", "license:apache-2.0", "8-bit", "exl2", "region:us" ]
null
2024-10-13T04:30:09Z
--- license: apache-2.0 --- # ArliAI-RPMax-12B-v1.2 ===================================== Saw ArliAI came out with a new version of the MN model and since I quanted it into EXL2 before figured I'd do it again. Though tbh I prefer Stardust v2, it's possible that this new update fixes the issues I had with it previously. Who knows? [This is the EXL2 8bpw version of this model. Find the original model here.](https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2) <br> [Find the 6bpw version here.](https://huggingface.co/Statuo/ArliAI-MN-12b-v1.2-EXL2-6bpw) <br> [Find the 4bpw version here.](https://huggingface.co/Statuo/ArliAI-MN-12b-v1.2-EXL2-4bpw) <br> ## UPDATE: For those getting gibberish results, it was merged wrongly to base after LORA training. Reuploaded all the files so it should work properly now. ## RPMax Series Overview | [2B](https://huggingface.co/ArliAI/Gemma-2-2B-ArliAI-RPMax-v1.1) | [3.8B](https://huggingface.co/ArliAI/Phi-3.5-mini-3.8B-ArliAI-RPMax-v1.1) | [8B](https://huggingface.co/ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.2) | [9B](https://huggingface.co/ArliAI/Gemma-2-9B-ArliAI-RPMax-v1.1) | [12B](https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2) | [20B](https://huggingface.co/ArliAI/InternLM2_5-20B-ArliAI-RPMax-v1.1) | [22B](https://huggingface.co/ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1) | [70B](https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1) | RPMax is a series of models that are trained on a diverse set of curated creative writing and RP datasets with a focus on variety and deduplication. This model is designed to be highly creative and non-repetitive by making sure no two entries in the dataset have repeated characters or situations, which makes sure the model does not latch on to a certain personality and be capable of understanding and acting appropriately to any characters or situations. Early tests by users mentioned that these models does not feel like any other RP models, having a different style and generally doesn't feel in-bred. You can access the model at https://arliai.com and ask questions at https://www.reddit.com/r/ArliAI/ We also have a models ranking page at https://www.arliai.com/models-ranking Ask questions in our new Discord Server! https://discord.com/invite/t75KbPgwhk ## Model Description ArliAI-RPMax-12B-v1.2 is a variant based on Mistral Nemo 12B Instruct 2407. This is arguably the most successful RPMax model due to how Mistral is already very uncensored in the first place. v1.2 update completely removes non-creative/RP examples in the dataset and is also an incremental improvement of the RPMax dataset which dedups the dataset even more and better filtering to cutout irrelevant description text that came from card sharing sites. ### Specs * **Context Length**: 128K * **Parameters**: 12B ### Training Details * **Sequence Length**: 8192 * **Training Duration**: Approximately 2 days on 2x3090Ti * **Epochs**: 1 epoch training for minimized repetition sickness * **LORA**: 64-rank 128-alpha, resulting in ~2% trainable weights * **Learning Rate**: 0.00001 * **Gradient accumulation**: Very low 32 for better learning. ## Quantization The model is available in quantized formats: * **FP16**: https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2 * **GGUF**: https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2-GGUF ## Suggested Prompt Format Mistral Instruct Prompt Format
josedossantos/urf-summary-labse
josedossantos
2024-10-13T04:55:34Z
7
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "sentence-similarity", "feature-extraction", "dataset_size:10K<n<100K", "loss:ContrastiveLoss", "arxiv:1908.10084", "base_model:sentence-transformers/LaBSE", "base_model:finetune:sentence-transformers/LaBSE", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T04:39:11Z
--- language: [] library_name: sentence-transformers tags: - sentence-transformers - sentence-similarity - feature-extraction - dataset_size:10K<n<100K - loss:ContrastiveLoss base_model: sentence-transformers/LaBSE widget: - source_sentence: ' O projeto de lei visa disciplinar os pedidos de vista nos tribunais brasileiros, estabelecendo prazos fixos e penalidades para a demora na restituição dos autos. O objetivo é impedir abusos e garantir uma julgamento justo e técnico. A proposta mantém a prerrogativa dos juízes de requerer vista dos autos, mas com uma disciplina mais rígida.' sentences: - ' O projeto de lei nº , de 2016, apresentado pelo deputado Diego Garcia, visa estabelecer um prazo de cinco sessões para o estudo dos pedidos de vistas nos tribunais, com o objetivo de tornar mais célere o julgamento de processos. A proposta se aplicará tanto ao processo penal quanto ao processo civil. A justificativa para a proposta é a necessidade de prevenir a morosidade nos julgamentos e a sensação de impunidade na sociedade, sem olvidar a existência do duplo grau de jurisdição.' - ' O projeto de lei visa estabelecer isenção de pedágio para veículos automotores de duas rodas, com o objetivo de melhorar a fluidez do trânsito e reduzir o risco de acidentes. A justificativa é que esses veículos não respondem por danos ao pavimento e não contribuem para o crescimento da demanda por espaço na via. Além disso, a cobrança de pedágio é difícil para esses veículos e pode levar a acidentes. A proposta foi apresentada pelo Deputado Hercílio Coelho e está sendo discutida na Câmara dos Deputados.' - ' O projeto de lei nº , de 2015, proposto pela deputada Mara Gabrilli, tipifica como crime o exercício ilegal das profissões de saúde de nível superior, incluindo as relacionadas pelo Conselho Nacional de Saúde. O projeto busca responder a situações em que profissionais sem autorização exercem as profissões de saúde, colocando em risco a saúde pública. A lei prevê penas para aqueles que exercem as profissões de saúde sem autorização legal ou excederem os limites de suas atribuições.' - source_sentence: ' O projeto de lei visa permitir que os contribuintes deduzam das receitas tributáveis as despesas realizadas com matrículas e mensalidades de cursos de idiomas estrangeiros, com o objetivo de incentivar a educação e o desenvolvimento profissional. A proposta foi apresentada pelo deputado Fernando Coruja e está sendo discutida no Congresso Nacional.' sentences: - ' O projeto de lei permite a dedução integral das despesas com educação do contribuinte e seus dependentes, incluindo cursos de línguas estrangeiras e aquisição de material didático e pedagógico, na apuração do Imposto de Renda da pessoa física. A justificativa é que a educação é um investimento escolhido pelos contribuintes e que essas despesas não são perdas para o Tesouro Nacional, mas sim um investimento estratégico para o desenvolvimento econômico do país. A proposta busca beneficiar a educação de todos os brasileiros e aumentar a competitividade do país no mercado global.' - ' O projeto de lei proíbe a fiança nos contratos de locação, quando o locatário for pessoa física e instituição religiosa, e permite a cobrança antecipada de até 3 (três) de aluguéis. Além disso, o projeto modifica a Lei 8.245/91, permitindo o pagamento antecipado do aluguel e estabelecendo que o locador é obrigado a pagar impostos e taxas. A proposta é justificada por causa das dificuldades que os pretendentes à locação enfrentam para encontrar um fiador proprietário e para evitar a penhora da única residência do fiador.' - ' O projeto de lei proposto por Laerte Bessa, PMDB-DF, visa regulamentar a jornada de trabalho de profissionais de saúde, estabelecendo um limite de 80 horas semanais para esses profissionários. A proposta é baseada na Constituição e na necessidade de garantir a qualidade do trabalho e a dignidade dos profissionais de saúde, que atualmente enfrentam demandas exageradas e desumanas. A medida busca evitar a contratação de servidores empregados em horários extras e rotineiros, garantindo o respeito ao limite de jornada semanal.' - source_sentence: ' O projeto de lei visa a tornar menos oneroso o cumprimento da sentença que reconhece a exigibilidade de obrigação de pagar quantia certa, diminui o valor da multa pelo não pagamento voluntário e diminui o valor da taxa dos juros a ser aplicada, com o objetivo de um maior equilíbrio processual entre as partes.' sentences: - ' O projeto de lei visa alterar a Lei nº 9.605/98 (Lei de Crimes Ambientais) para dispor sobre o valor das multas em caso de desastre ambiental. O objetivo é flexibilizar a Lei, permitindo que o valor da multa seja aumentado em até cem vezes do teto, em caso de desastre ambiental, proporcionalmente ao nível dos danos causados à saúde humana ou ao meio ambiente. Além disso, o projeto estabelece que o infrator deve reparar os danos causados, independentemente de se optar por converter a multa em serviços de preservação e melhoria do meio ambiente.' - ' O projeto de lei visa regular o inciso LI, art. 5º, da Constituição Federal, que estabelece a possibilidade de extradição de brasileiros naturalizados, quando forem condenados por crimes cometidos antes da naturalização. O projeto busca preencher a lacuna existente na legislação e garantir a cooperação internacional no combate aos crimes transnacionais. Além disso, o projeto estabelece que a extradição de brasileiros naturalizados será possível apenas quando houver comprovado envolvimento do réu na prática do crime.' - ' O projeto de lei visa estabelecer a duração semanal de trabalho para enfermeiros técnicos, enfermagem auxiliares e enfermagem parteiras, que atualmente é regulada pela Lei nº 7.498/1986. A proposta é de fixar a jornada diária em 30 horas, com o objetivo de garantir um descanso adequado para os profissionais da saúde e evitar a fadiga e o desgaste. A proposta também busca estabelecer a duração semanal de trabalho para esses profissionais em 30 horas, com o objetivo de garantir um descanso adequado para os profissionais da saúde e evitar a fadiga e o desgaste.' - source_sentence: ' O projeto de lei proíbe a realização de chamadas telefônicas anônimas, obrigando a identificação do remetente e do destinatário. A justificativa é a necessidade de combater crimes e violações de direitos garantidos, que ocorrem por meio do uso inadequado da telefonia. A proibição será aplicada a partir da data de publicação da lei.' sentences: - ' O projeto de lei proíbe a realização de ligações telefônicas não identificáveis, com o objetivo de inibir a proliferação de ligações confidenciais com fins de telemarketing e cobrança, e também para reduzir os índices de criminalidade no país. O projeto mantém o dispositivo da Lei Geral de Telecomunicações que assegura ao usuário o direito à não divulgação do seu número telefônico, mas apenas em listas de assinantes. O projeto estabelece um prazo de noventa dias para que a norma proposta passe a vigorar, permitindo que as operadoras adaptem suas redes para adequação ao disposto na proposição.' - ' O Projeto de Lei nº 9.394/96, apresentado pela Deputada Federal Kátia Abreu em 2005, visa estabelecer diretrizes para a educação nacional e instituir o Programa Educação Cidadania (PEC). O projeto prevê a implantação do ensino médio em escolas públicas, com foco em estudos organização política administrativa e cidadania, e estabelece prazos para a implantação em diferentes estados e municípios. Além disso, o projeto busca popularizar o conhecimento da divisão política administrativa do país e da função da cidadania.' - ' O projeto de lei de 2008 da Deputada Solange Almeida (PMDB-RJ) dispõe desconto de 35% nas taxas de inscrição para candidatos que atendam a critérios de deficiência física, necessidades especiais e renda mensal inferior a dois salários mínimos. A lei também estabelece que os doadores de sangue possam receber um desconto de 35% nas taxas de inscrição. Além disso, a lei prevê que as informações estabelecidas no artigo 1º sejam incluídas nos editais dos concursos públicos. A lei entrará em vigor 180 dias após a publicação oficial.' - source_sentence: ' O projeto de lei reconhece o Carnaval do Município de Aracati-CE como manifestação da cultura nacional, destacando sua importância econômica e social. O evento é realizado em quatro arenas e atrai turistas de todo o país. A cidade de Aracati preserva suas raízes e tradições, e o projeto pode contribuir para a perpetuação da cultura do povo.' sentences: - ' O projeto de lei reconhece o Carnaval de Nova Russas como manifestação da cultura nacional e incentiva a valorização e ampliação das manifestações culturais do Brasil. A celebração é uma importante fonte de renda econômica para o município e atrai turistas de todas as regiões. A lei foi criada em 2017 e incluiu o Carnaval de Nova Russas no Calendário Turístico Cultural do Estado do Ceará. O projeto tem importante relevância cultural e social e conta com o apoio dos ilustres pares para sua aprovação.' - ' O projeto de lei do deputado Eliene Lima visa modificar o Decreto Lei 5.518/2005, que estabelece a admissão automática de diplomas de pós-graduação expedidos por universidades e faculdades estrangeiras para o mercado brasileiro. O projeto busca garantir a qualidade do ensino e a proteção dos jovens cidadãos, evitando a entrada de diplomas de qualidade duvidosa no mercado. Além disso, o projeto também busca preservar a soberania interna dos Estados Integrantes do Mercosul e evitar a ultrapassagem de limites legais.' - ' O projeto de lei visa alterar o art. 282 do Decreto Lei 2.848/40, permitindo que os técnicos tecnólogos radiologia possam exercer as profissões de médico, dentista e farmacêutico, sem a necessidade de autorização legal e ética, e aumentando os limites de responsabilidade penal para essas atividades. O projeto também busca inserir condutas previstas na legislação para evitar o exercício ilegal da profissão e minimizar os efeitos nocivos para os seres vivos e a sociedade.' pipeline_tag: sentence-similarity --- # SentenceTransformer based on sentence-transformers/LaBSE This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/LaBSE](https://huggingface.co/sentence-transformers/LaBSE). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [sentence-transformers/LaBSE](https://huggingface.co/sentence-transformers/LaBSE) <!-- at revision e34fab64a3011d2176c99545a93d5cbddc9a91b7 --> - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 768 tokens - **Similarity Function:** Cosine Similarity <!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown --> ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("josedossantos/urf-summary-labse") # Run inference sentences = [ ' O projeto de lei reconhece o Carnaval do Município de Aracati-CE como manifestação da cultura nacional, destacando sua importância econômica e social. O evento é realizado em quatro arenas e atrai turistas de todo o país. A cidade de Aracati preserva suas raízes e tradições, e o projeto pode contribuir para a perpetuação da cultura do povo.', ' O projeto de lei reconhece o Carnaval de Nova Russas como manifestação da cultura nacional e incentiva a valorização e ampliação das manifestações culturais do Brasil. A celebração é uma importante fonte de renda econômica para o município e atrai turistas de todas as regiões. A lei foi criada em 2017 e incluiu o Carnaval de Nova Russas no Calendário Turístico Cultural do Estado do Ceará. O projeto tem importante relevância cultural e social e conta com o apoio dos ilustres pares para sua aprovação.', ' O projeto de lei do deputado Eliene Lima visa modificar o Decreto Lei 5.518/2005, que estabelece a admissão automática de diplomas de pós-graduação expedidos por universidades e faculdades estrangeiras para o mercado brasileiro. O projeto busca garantir a qualidade do ensino e a proteção dos jovens cidadãos, evitando a entrada de diplomas de qualidade duvidosa no mercado. Além disso, o projeto também busca preservar a soberania interna dos Estados Integrantes do Mercosul e evitar a ultrapassagem de limites legais.', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 768] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` <!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* --> ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 10,962 training samples * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code> * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | label | |:--------|:-------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 46 tokens</li><li>mean: 138.47 tokens</li><li>max: 377 tokens</li></ul> | <ul><li>min: 55 tokens</li><li>mean: 136.56 tokens</li><li>max: 358 tokens</li></ul> | <ul><li>0: ~49.70%</li><li>1: ~50.30%</li></ul> | * Samples: | sentence_0 | sentence_1 | label | |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------| | <code> O projeto de lei proíbe a realização de ligações telefônicas não identificáveis, com o objetivo de inibir a proliferação de ligações confidenciais com fins de telemarketing e cobrança, e também para reduzir os índices de criminalidade no país. O projeto mantém o dispositivo da Lei Geral de Telecomunicações que assegura ao usuário o direito à não divulgação do seu número telefônico, mas apenas em listas de assinantes. O projeto estabelece um prazo de noventa dias para que a norma proposta passe a vigorar, permitindo que as operadoras adaptem suas redes para adequação ao disposto na proposição.</code> | <code> O projeto de lei visa obrigar as operadoras de telefonia fixa e celular a informarem o código de acesso do usuário chamador em ligações telefônicas, sem ônus para o usuário receptor. A medida busca reforçar a segurança pública e facilitar as investigações e a prisão de criminosos que se escondem atrás da situação anônima da chamada não identificada. O projeto foi apresentado pelo deputado federal Alberto Fraga e justifica-se com a necessidade de combater a criminalidade que utiliza as telecomunicações para perpetrar crimes.</code> | <code>1</code> | | <code> O projeto de lei nº 429/2020, apresentado pela deputada Tabata Amaral, visa permitir a ausência do empregado ao trabalho, sem prejuízo do salário, para participar de reuniões escolares de seus dependentes. A proposta busca contribuir para uma educação de qualidade no Brasil, garantindo o direito da criança, do adolescente e do jovem aos direitos fundamentais, como estabelecido na Constituição Federal de 1988. Ao permitir a ausência dos empregados para participar de reuniões escolares, a proposta busca aumentar a participação dos pais na evolução de seus filhos nas escolas, contribuindo para a prevenção de bullying e para o senso de valor e na auto-percepção da dignidade das crianças. Além disso, a proposta busca incentivar a soma dos esforços de todas as esferas para garantir os direitos fundamentais da criança, do adolescente e do jovem.</code> | <code> O projeto de lei proposto por Ricardo Izar PSD/SP, dispõe sobre o acompanhamento pedagógico de crianças e adolescentes na escola pelos pais ou responsáveis legais. O projeto alterará três leis importantes: Estatuto da Criança e do Adolescente, Lei de Diretrizes e Bases da Educação Nacional e Consolidação das Leis do Trabalho. O objetivo é contribuir para a promoção do acompanhamento efetivo e sistemático familiar da vida escolar dos filhos, para que os pais possam participar de forma mais ativa da educação de seus filhos. A proposta tem como base a ideia de que a participação dos pais na vida escolar dos filhos é fundamental para o seu desenvolvimento e aprendizado.</code> | <code>1</code> | | <code> o projeto de lei visa estabelecer regras para as campanhas de telemarketing ativo, com foco em garantir taxa máxima de abandono de ligações, evitar insistência nas campanhas e garantir o direito do consumidor de não ser importunado novamente por determinado período. Além disso, o projeto propõe a inserção de um novo artigo no código de defesa do consumidor para controlar o uso agressivo e indiscriminado dos discadores preditivos.</code> | <code> O projeto de lei visa criar o Sistema Nacional de Bloqueio de Telemarketing, com o objetivo de proteger os usuários dos serviços de telefonia de chamadas indesejadas. O sistema será de acesso livre e gratuito por qualquer usuário dos serviços de telecomunicações e estarão disponíveis 24 horas por dia e 7 dias por semana. As empresas que executam atividade de atendimento ativo para consumidores por meio de serviço de telecomunicações terão obrigação de disponibilizar serviços de atendimento à distância para o acesso ao Sistema Nacional de Bloqueio de Telemarketing. A falta de cumprimento da norma pode sujeitar o infrator à pena de multa. A justificação do projeto é que as atividades de telemarketing estão cada vez mais presentes no Brasil e que o abuso nesse uso vem causando grande insatisfação aos cidadãos.</code> | <code>0</code> | * Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters: ```json { "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", "margin": 0.5, "size_average": true } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters <details><summary>Click to expand</summary> - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin </details> ### Training Logs | Epoch | Step | Training Loss | |:------:|:----:|:-------------:| | 0.0912 | 500 | 0.0266 | | 0.1824 | 1000 | 0.0252 | | 0.2737 | 1500 | 0.0221 | | 0.3649 | 2000 | 0.0205 | | 0.4561 | 2500 | 0.0189 | | 0.5473 | 3000 | 0.0169 | | 0.6386 | 3500 | 0.0156 | | 0.7298 | 4000 | 0.0132 | | 0.8210 | 4500 | 0.0138 | | 0.9122 | 5000 | 0.0123 | ### Framework Versions - Python: 3.10.14 - Sentence Transformers: 3.0.0 - Transformers: 4.39.3 - PyTorch: 2.2.0 - Accelerate: 0.30.1 - Datasets: 2.14.4 - Tokenizers: 0.15.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### ContrastiveLoss ```bibtex @inproceedings{hadsell2006dimensionality, author={Hadsell, R. and Chopra, S. and LeCun, Y.}, booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, title={Dimensionality Reduction by Learning an Invariant Mapping}, year={2006}, volume={2}, number={}, pages={1735-1742}, doi={10.1109/CVPR.2006.100} } ``` <!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->
Arkana08/NIHAPPY-L3.1-8B-v0.09
Arkana08
2024-10-13T04:55:16Z
12
7
transformers
[ "transformers", "safetensors", "llama", "text-generation", "mergekit", "merge", "conversational", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-01T06:59:14Z
--- base_model: [] library_name: transformers tags: - mergekit - merge --- ### (GGUF) Thanks: #### mradermacher - **GGUF:** **[mradermacher/NIHAPPY-L3.1-8B-v0.09-GGUF](https://huggingface.co/mradermacher/NIHAPPY-L3.1-8B-v0.09-GGUF)** - **imatrix GGUF:** **[mradermacher/NIHAPPY-L3.1-8B-v0.09-i1-GGUF](https://huggingface.co/mradermacher/NIHAPPY-L3.1-8B-v0.09-i1-GGUF)** #### QuantFactory - **GGUF:** **[QuantFactory/NIHAPPY-L3.1-8B-v0.09-GGUF](https://huggingface.co/QuantFactory/NIHAPPY-L3.1-8B-v0.09-GGUF)** #### RichardErkhov - **GGUF:** **[RichardErkhov/Arkana08_-_NIHAPPY-L3.1-8B-v0.09-gguf](https://huggingface.co/RichardErkhov/Arkana08_-_NIHAPPY-L3.1-8B-v0.09-gguf)** **NIHAPPY** a role-playing model that seamlessly integrates the finest qualities of various pre-trained language models. Crafted with a focus on dynamic storytelling, NIHAPPY thrives in creative and immersive roleplaying environments. It masterfully balances intricate reasoning with a smooth and consistent narrative flow, ensuring an engaging and coherent experience. ## Merge Details ### Merge Method This model was created using the **DARE TIES** merge method, with **[v000000/L3.1-Niitorm-8B-DPO-t0.0001](https://huggingface.co/v000000/L3.1-Niitorm-8B-DPO-t0.0001)** serving as the base model. This approach ensures a fine-tuned balance between creativity, reasoning, and rule-following, ideal for intricate role-playing situations. ### Models Merged The following models were included in the merge: - **[Nitral-AI/Hathor_Tahsin-L3-8B-v0.85](https://huggingface.co/Nitral-AI/Hathor_Tahsin-L3-8B-v0.85)**: Contributes to rule-following, consistency, and character accuracy. - **[v000000/L3.1-Niitorm-8B-DPO-t0.0001](https://huggingface.co/v000000/L3.1-Niitorm-8B-DPO-t0.0001)**: A strong foundation for reasoning and logical responses. - **[v000000/L3-8B-Poppy-Moonfall-C](https://huggingface.co/v000000/L3-8B-Poppy-Moonfall-C)**: Focuses on storytelling, imagination, and creative expression. ## Configuration The following YAML configuration was used to produce NIHAPPY: ```yaml models: - model: v000000/L3.1-Niitorm-8B-DPO-t0.0001 parameters: weight: 0.35 density: 0.65 - model: v000000/L3-8B-Poppy-Moonfall-C parameters: weight: 0.45 density: 0.65 - model: Nitral-AI/Hathor_Tahsin-L3-8B-v0.85 parameters: weight: 0.20 density: 0.68 merge_method: dare_ties base_model: v000000/L3.1-Niitorm-8B-DPO-t0.0001 parameters: int8_mask: true dtype: bfloat16 ``` ## Credits Thanks to the creators of the models: - **[Nitral-AI/Hathor_Tahsin-L3-8B-v0.85](https://huggingface.co/Nitral-AI/Hathor_Tahsin-L3-8B-v0.85)** - **[v000000/L3.1-Niitorm-8B-DPO-t0.0001](https://huggingface.co/v000000/L3.1-Niitorm-8B-DPO-t0.0001)** - **[v000000/L3-8B-Poppy-Moonfall-C](https://huggingface.co/v000000/L3-8B-Poppy-Moonfall-C)**
ammonbro/something
ammonbro
2024-10-13T04:42:50Z
177
0
transformers
[ "transformers", "safetensors", "gpt2", "text-generation", "generated_from_trainer", "base_model:distilbert/distilgpt2", "base_model:finetune:distilbert/distilgpt2", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T04:35:27Z
--- library_name: transformers license: apache-2.0 base_model: distilgpt2 tags: - generated_from_trainer model-index: - name: something results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # something This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.7061 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:-----:|:---------------:| | 0.6969 | 1.0 | 1429 | 0.7070 | | 0.6916 | 2.0 | 2858 | 0.7016 | | 0.6905 | 3.0 | 4287 | 0.7114 | | 0.6887 | 4.0 | 5716 | 0.7091 | | 0.6888 | 5.0 | 7145 | 0.7066 | | 0.6878 | 6.0 | 8574 | 0.7043 | | 0.6874 | 7.0 | 10003 | 0.7076 | | 0.6877 | 8.0 | 11432 | 0.7072 | | 0.6868 | 9.0 | 12861 | 0.7067 | | 0.6869 | 10.0 | 14290 | 0.7061 | ### Framework versions - Transformers 4.45.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.20.1
josedossantos/urf-txtIndexacao-legalbertpt
josedossantos
2024-10-13T04:39:03Z
5
1
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "sentence-similarity", "feature-extraction", "dataset_size:10K<n<100K", "loss:ContrastiveLoss", "arxiv:1908.10084", "base_model:raquelsilveira/legalbertpt_fp", "base_model:finetune:raquelsilveira/legalbertpt_fp", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T04:35:16Z
--- language: [] library_name: sentence-transformers tags: - sentence-transformers - sentence-similarity - feature-extraction - dataset_size:10K<n<100K - loss:ContrastiveLoss base_model: raquelsilveira/legalbertpt_fp widget: - source_sentence: Alteração, fixação, jornada de trabalho, psicólogo. sentences: - "Alteração, lei federal, definição, jornada de trabalho, psicólogo.\r\n\r\n" - Concessão, Pensão especial, pessoa, Sequela, Coronavírus, sujeição, tratamento médico, ineficácia, diretrizes. - Alteração, Código Civil, garantia, companheiro, direito real, habitação, imóvel residencial, inventário. - source_sentence: Criação, Fundo Garantidor, empresa, alimentação. sentences: - 'Critérios, concessão, auxíio financeiro, Municípios, compensação, redução, cota, Fundo de Participação dos Municípios (FPM). ' - Alteração, Lei dos Crimes Hediondos, inclusão, crime hediondo, concussão, corrupção ativa, corrupção passiva. - Constituição federal (1988), Direitos e garantias fundamentais, acesso, Internet, inviolabilidade, sigilo, comunicação eletrônica. - source_sentence: Fixação, preço, Gás Liquefeito de Petróleo (GLP). sentences: - Autorização, Porto do Forno, município, Arraial do Cabo, (RJ), importação, exportação, biocombustível. - 'Obrigatoriedade, instalação, agência lotérica, banheiro feminino, banheiro masculino, bebedouro, consumidor. ' - Proibição, empresa, telefonia móvel, mensagem, cobrança, inadimplência, ligação, cliente. - source_sentence: Fixação, prazo, mandato, membro, Tribunal de Contas. sentences: - 'Constituição Federal (1988), criação, mandato coletivo, mandato parlamentar. ' - 'Alteração, Lei Antifumo, teor alcóolico, proibição, propaganda comercial, bebida alcoólica, comunicação de massa. ' - Obrigatoriedade, restaurante, concessão, desconto, cliente, cirurgia bariátrica, gastroplastia endoscópica, descumprimento, multa. - source_sentence: Regulamentação, profissão, designer de interiores. sentences: - Regulamentação profissional, Influenciador digital, criação, geração, Conteúdo digital, Rede social, Mídia social, atribuição, deveres. - 'Proibição, nomeação, homem, Cargo em comissão, Administração federal, condenação, crime, violência contra mulher. ' - 'Alteração, Código Penal, crime contra a liberdade sexual, tipicidade penal, violação sexual mediante fraude, utilização, sedação, reclusão. ' pipeline_tag: sentence-similarity --- # SentenceTransformer based on raquelsilveira/legalbertpt_fp This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [raquelsilveira/legalbertpt_fp](https://huggingface.co/raquelsilveira/legalbertpt_fp). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [raquelsilveira/legalbertpt_fp](https://huggingface.co/raquelsilveira/legalbertpt_fp) <!-- at revision c6d8158c5561e78815d354efce6ff77a9e6730c7 --> - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 768 tokens - **Similarity Function:** Cosine Similarity <!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown --> ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("josedossantos/urf-txtIndexacao-legalbertpt") # Run inference sentences = [ 'Regulamentação, profissão, designer de interiores.', 'Regulamentação profissional, Influenciador digital, criação, geração, Conteúdo digital, Rede social, Mídia social, atribuição, deveres.', 'Proibição, nomeação, homem, Cargo em comissão, Administração federal, condenação, crime, violência contra mulher. ', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 768] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` <!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* --> ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 10,962 training samples * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code> * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | label | |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 9 tokens</li><li>mean: 49.68 tokens</li><li>max: 249 tokens</li></ul> | <ul><li>min: 9 tokens</li><li>mean: 53.11 tokens</li><li>max: 421 tokens</li></ul> | <ul><li>0: ~49.90%</li><li>1: ~50.10%</li></ul> | * Samples: | sentence_0 | sentence_1 | label | |:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------| | <code>Alteração, Lei de Benefícios da Previdência Social, criação, disciplinamento, auxílio-cuidador, segurado, Regime Geral de Previdência Social (RGPS), familiar, exercício, atividade, cuidador de deficientes.</code> | <code>Alteração, Estatuto do Idoso, requisito, exercício profissional, cuidador de idosos. _Poder público, estímulo, adoção, idoso, campanha educativa. </code> | <code>1</code> | | <code>Equiparação, doença, Lúpus Eritematoso Sistêmico, deficiência física, deficiência intelectual, efeito jurídico.</code> | <code>Criação, Política Nacional de Conscientização e Orientação sobre LES, combate, doença grave, campanha educativa, tratamento médico, informações, coleta, dados, portador, doença, pesquisa científica, garantia, acesso, medicamentos, inclusão, cosméticos, bloqueador solar, proteção, radiação ultravioleta, pele.</code> | <code>0</code> | | <code>Alteração, Lei de Isenção do IPI para Compra de Automóveis, critério, isenção tributária, Imposto sobre Produtos Industrializados (IPI), aquisição, Automóvel, motorista, Transporte individual, transporte de passageiro, Motorista de aplicativo, benefício fiscal, tributação.</code> | <code>Alteração, Lei de Isenção do IPI para Compra de Automóveis, isenção, Imposto sobre Produtos Industrializados (IPI), motorista de aplicativo, aquisição, veículo de passageiro, tributação.</code> | <code>1</code> | * Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters: ```json { "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", "margin": 0.5, "size_average": true } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters <details><summary>Click to expand</summary> - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin </details> ### Training Logs | Epoch | Step | Training Loss | |:------:|:----:|:-------------:| | 0.0912 | 500 | 0.0278 | | 0.1824 | 1000 | 0.0242 | | 0.2737 | 1500 | 0.0226 | | 0.3649 | 2000 | 0.0201 | | 0.4561 | 2500 | 0.0189 | | 0.5473 | 3000 | 0.0165 | | 0.6386 | 3500 | 0.0148 | | 0.7298 | 4000 | 0.0135 | | 0.8210 | 4500 | 0.0122 | | 0.9122 | 5000 | 0.0128 | ### Framework Versions - Python: 3.10.14 - Sentence Transformers: 3.0.0 - Transformers: 4.39.3 - PyTorch: 2.2.0 - Accelerate: 0.30.1 - Datasets: 2.14.4 - Tokenizers: 0.15.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### ContrastiveLoss ```bibtex @inproceedings{hadsell2006dimensionality, author={Hadsell, R. and Chopra, S. and LeCun, Y.}, booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, title={Dimensionality Reduction by Learning an Invariant Mapping}, year={2006}, volume={2}, number={}, pages={1735-1742}, doi={10.1109/CVPR.2006.100} } ``` <!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->
josedossantos/urf-txtIndexacao-legalbertimbau
josedossantos
2024-10-13T04:35:13Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "sentence-similarity", "feature-extraction", "dataset_size:10K<n<100K", "loss:ContrastiveLoss", "arxiv:1908.10084", "base_model:rufimelo/Legal-BERTimbau-large", "base_model:finetune:rufimelo/Legal-BERTimbau-large", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T04:23:31Z
--- language: [] library_name: sentence-transformers tags: - sentence-transformers - sentence-similarity - feature-extraction - dataset_size:10K<n<100K - loss:ContrastiveLoss base_model: rufimelo/Legal-BERTimbau-large widget: - source_sentence: Critério, cobrança, Taxa de juros, Cheque especial sentences: - 'Critério, limitação, Taxa de juros, cheque especial, cartão de crédito, Instituição Financeira. ' - Alteração, Lei Antifumo, teor alcoólico, bebida alcoólica, propaganda comercial, rótulo, embalagem, advertência. - Alteração, Código Penal, tipicidade, crime, exercício ilegal da profissão, tecnólogo, técnico em radiologia, falta, habilitação profissional. - source_sentence: Inclusão, Cerrado, Caatinga, Patrimônio da União. sentences: - Inclusão, cerrado, caatinga, patrimônio da União. - 'Título de Capital Nacional, Capital Nacional do Alimento, Marília (SP), São Paulo (Estado), Título de Topônimo. ' - 'Proibição, Robocalls, Empresa de call center, telemarketing, procedimento, ligação telefônica, infração, multa. ' - source_sentence: Regulamentação, profissão, Terapeuta Ocupacional. sentences: - 'Alteração, Lei Orgânica da Saúde, Assistência terapêutica integral, critério, Dispensação de medicamento. ' - Alteração, Código de Processo Civil, critérios, adjudicação, bens imóveis, autorização, aplicação, bens móveis. - Alteração, Código Penal, tipicidade penal, desrespeito à ordem de vacinação, violação, ordem de prioridade, vacinação. - source_sentence: Isenção, taxa, pedágio, rodovia federal, motocicleta. sentences: - Alteração, decreto-lei federal, concessão, isenção, pagamento, pedágio, motocicleta, ciclomotor, triciclo. - "Suspensão, cobrança, juros, cheque especial, cartão de crédito, pandemia, coronavírus.\ \ Emergência social.\r\n\r\n" - Oferta, Absorvente feminino, unidade de saúde, Atenção primária à saúde. - source_sentence: Criação, Dia Nacional, Verdade, comemoração, outubro. sentences: - 'Criação, Dia Nacional do Perito Odontolegista, comemoração, setembro. ' - 'Alteração, Lei de Participação nos Lucros das Empresas, garantia, autorização, abertura, comércio, domingo, feriado. ' - 'Alteração, Estatuto do Idoso, requisito, exercício profissional, cuidador de idosos. _Poder público, estímulo, adoção, idoso, campanha educativa. ' pipeline_tag: sentence-similarity --- # SentenceTransformer based on rufimelo/Legal-BERTimbau-large This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [rufimelo/Legal-BERTimbau-large](https://huggingface.co/rufimelo/Legal-BERTimbau-large). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [rufimelo/Legal-BERTimbau-large](https://huggingface.co/rufimelo/Legal-BERTimbau-large) <!-- at revision 639633891a335ae765de5746673563b060f06488 --> - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 1024 tokens - **Similarity Function:** Cosine Similarity <!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown --> ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("josedossantos/urf-txtIndexacao-legalbertimbau") # Run inference sentences = [ 'Criação, Dia Nacional, Verdade, comemoração, outubro.', 'Criação, Dia Nacional do Perito Odontolegista, comemoração, setembro. ', 'Alteração, Lei de Participação nos Lucros das Empresas, garantia, autorização, abertura, comércio, domingo, feriado. ', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 1024] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` <!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* --> ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 10,962 training samples * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code> * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | label | |:--------|:------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 10 tokens</li><li>mean: 51.67 tokens</li><li>max: 402 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 53.91 tokens</li><li>max: 402 tokens</li></ul> | <ul><li>0: ~51.50%</li><li>1: ~48.50%</li></ul> | * Samples: | sentence_0 | sentence_1 | label | |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------| | <code>Alteração, prorrogação, prazo, anistia, estrangeiro, ilegalidade, País, Brasil, requerimento, órgão público, registro provisório, visto de passaporte, regularização, permanência, território nacional. <br></code> | <code>Alteração, Lei de Anistia ao Servidor Público, inclusão, beneficiário, anistia, ex-servidor, empregado, órgão público, empresa estatal, vinculação, Administração Pública.</code> | <code>0</code> | | <code>Alteração, Código Civil (2002), extinção, cobrança, laudêmio, transmissão, bem aforado.</code> | <code>Alteração, Código Civil, autorização, cobrança, laudêmio, transmissão, bens, aforamento, aplicação de recursos, manutenção, entidade, instituição assistencial, instituição religiosa, obra filantrópica.</code> | <code>1</code> | | <code>Alteração, Código Penal, crime contra a liberdade sexual, tipicidade penal, violação sexual mediante fraude, utilização, sedação, reclusão. </code> | <code>Alteração, Lei Caó, crime, fabricação, comercialização, distribuição, símbolo, emblema, divulgação, nazismo, fascismo, racismo, xenofobia, separatismo.</code> | <code>0</code> | * Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters: ```json { "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", "margin": 0.5, "size_average": true } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters <details><summary>Click to expand</summary> - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin </details> ### Training Logs | Epoch | Step | Training Loss | |:------:|:----:|:-------------:| | 0.0912 | 500 | 0.033 | | 0.1824 | 1000 | 0.0247 | | 0.2737 | 1500 | 0.0227 | | 0.3649 | 2000 | 0.0194 | | 0.4561 | 2500 | 0.0177 | | 0.5473 | 3000 | 0.0173 | | 0.6386 | 3500 | 0.0138 | | 0.7298 | 4000 | 0.0127 | | 0.8210 | 4500 | 0.0111 | | 0.9122 | 5000 | 0.0107 | ### Framework Versions - Python: 3.10.14 - Sentence Transformers: 3.0.0 - Transformers: 4.39.3 - PyTorch: 2.2.0 - Accelerate: 0.30.1 - Datasets: 2.14.4 - Tokenizers: 0.15.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### ContrastiveLoss ```bibtex @inproceedings{hadsell2006dimensionality, author={Hadsell, R. and Chopra, S. and LeCun, Y.}, booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, title={Dimensionality Reduction by Learning an Invariant Mapping}, year={2006}, volume={2}, number={}, pages={1735-1742}, doi={10.1109/CVPR.2006.100} } ``` <!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->
josedossantos/urf-txtIndexacao-bertimbau
josedossantos
2024-10-13T04:19:34Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "sentence-similarity", "feature-extraction", "dataset_size:10K<n<100K", "loss:ContrastiveLoss", "arxiv:1908.10084", "base_model:neuralmind/bert-large-portuguese-cased", "base_model:finetune:neuralmind/bert-large-portuguese-cased", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T04:08:18Z
--- language: [] library_name: sentence-transformers tags: - sentence-transformers - sentence-similarity - feature-extraction - dataset_size:10K<n<100K - loss:ContrastiveLoss base_model: neuralmind/bert-large-portuguese-cased widget: - source_sentence: Alteração, inclusão, evento esportivo, Vale-Cultura. sentences: - Criação, incentivo, atividade cultural, desconto, categoria profissional, professor, escola pública, parceria, cinema, teatro, editora, livraria. - 'Alteração, Lei das Eleições (1997), nome, registro de candidatura, candidato a cargo eletivo, justiça eleitoral. ' - Alteração, Lei do Marco Civil da Internet, proibição, provedor de acesso, Internet, cobrança, franquia, banda larga fixa. - source_sentence: Inclusão, Cerrado, Caatinga, Patrimônio da União. sentences: - Alteração, Constituição Federal, Meio Ambiente, inclusão, ecossistema, mar, caatinga, campo, Região Sul, Patrimônio da União. - Título de Capital Nacional, Capital Nacional do Alimento, Marília (SP), São Paulo (Estado), Título de Topônimo. - 'Proibição, cobrança, tarifa, conta bancária, inatividade, descumprimento, penalidade administrativa. ' - source_sentence: Criação, Semana Nacional dos Contadores de História. sentences: - Criação, Dia Nacional da Filantropia, comemoração, agosto. - 'Alteração, Lei Antifumo, teor alcóolico, proibição, propaganda comercial, bebida alcoólica, comunicação de massa. ' - Alteração, Lei da TV Paga, concessão, prestação de serviços, televisão por assinatura, possibilidade, adaptação, radiodifusão de sons e imagens. - source_sentence: Alteração, fixação, jornada de trabalho, psicólogo. sentences: - Alteração, regulamentação, jornada de trabalho, psicólogo. - Regulamentação profissional, garçom, exercício profissional, documentação, piso salarial, jornada de trabalho. - Alteração, lei federal, piso salarial, profissão, enfermeiro, técnico de enfermagem, auxiliar de enfermagem, obstetriz. - source_sentence: Voto facultativo, eleitor, maior de dezesseis anos. sentences: - Constituição Federal (1988), facultatividade, direito de voto, eleições, voto facultativo. - Regulamentação profissional, garçom, exercício profissional, documentação, piso salarial, jornada de trabalho. - "Alteração, Lei dos Direitos Autorais, suspensão, renovação, serviço, radiodifusão,\ \ rádio, inadimplência, direito autoral.\r\n" pipeline_tag: sentence-similarity --- # SentenceTransformer based on neuralmind/bert-large-portuguese-cased This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased) <!-- at revision aa302f6ea73b759f7df9cad58bd272127b67ec28 --> - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 1024 tokens - **Similarity Function:** Cosine Similarity <!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown --> ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("josedossantos/urf-txtIndexacao-bertimbau") # Run inference sentences = [ 'Voto facultativo, eleitor, maior de dezesseis anos.', 'Constituição Federal (1988), facultatividade, direito de voto, eleições, voto facultativo.', 'Regulamentação profissional, garçom, exercício profissional, documentação, piso salarial, jornada de trabalho.', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 1024] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` <!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* --> ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 10,962 training samples * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code> * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | label | |:--------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 9 tokens</li><li>mean: 52.32 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 52.76 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>0: ~49.90%</li><li>1: ~50.10%</li></ul> | * Samples: | sentence_0 | sentence_1 | label | |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------|:---------------| | <code>Alteração, Lei do Saneamento Básico, isenção, cobrança, utilização, recursos hídricos, Planta de dessalinização, água do mar, água salobra, abastecimento de água. <br>_ Política Federal de Saneamento Básico, União, incentivo, dessalinização, água do mar, água salobra, abastecimento de água. <br>_Alteração, Lei do Setor Elétrico, desconto, tarifa, energia elétrica, planta de dessalinização, água do mar, água salobra. <br></code> | <code>Exclusão, custos, transmissão, energia elétrica, consumidor, municípios, hidrelétrica.</code> | <code>0</code> | | <code>Definição, grau de insalubridade, atividade profissional, coleta, lixo, lixeiro, gari, garantia, aposentadoria especial.</code> | <code>Criação, Dia do Gari, comemoração, maio.</code> | <code>0</code> | | <code>Alteração, Lei do Setor Elétrico, desconto, tarifa, energia elétrica, consumo de energia, atividade, dessalinização, água salgada. <br> <br></code> | <code>Isenção, tarifa, energia elétrica, poço artesiano, abastecimento de água, consumo humano, animal, irrigação.</code> | <code>0</code> | * Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters: ```json { "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", "margin": 0.5, "size_average": true } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters <details><summary>Click to expand</summary> - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin </details> ### Training Logs | Epoch | Step | Training Loss | |:------:|:----:|:-------------:| | 0.0912 | 500 | 0.0328 | | 0.1824 | 1000 | 0.0238 | | 0.2737 | 1500 | 0.0206 | | 0.3649 | 2000 | 0.0182 | | 0.4561 | 2500 | 0.0165 | | 0.5473 | 3000 | 0.013 | | 0.6386 | 3500 | 0.0134 | | 0.7298 | 4000 | 0.0112 | | 0.8210 | 4500 | 0.0111 | | 0.9122 | 5000 | 0.0107 | ### Framework Versions - Python: 3.10.14 - Sentence Transformers: 3.0.0 - Transformers: 4.39.3 - PyTorch: 2.2.0 - Accelerate: 0.30.1 - Datasets: 2.14.4 - Tokenizers: 0.15.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### ContrastiveLoss ```bibtex @inproceedings{hadsell2006dimensionality, author={Hadsell, R. and Chopra, S. and LeCun, Y.}, booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, title={Dimensionality Reduction by Learning an Invariant Mapping}, year={2006}, volume={2}, number={}, pages={1735-1742}, doi={10.1109/CVPR.2006.100} } ``` <!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->
Priyasi/Pretrain-virusmodel_5
Priyasi
2024-10-13T04:17:56Z
7
0
null
[ "tensorboard", "safetensors", "bert", "custom_code", "license:apache-2.0", "region:us" ]
null
2024-10-08T09:58:21Z
--- license: apache-2.0 ---
wyb1001/sd-class-butterflies-64
wyb1001
2024-10-13T04:17:10Z
44
0
diffusers
[ "diffusers", "safetensors", "pytorch", "unconditional-image-generation", "diffusion-models-class", "license:mit", "diffusers:DDPMPipeline", "region:us" ]
unconditional-image-generation
2024-10-13T04:16:51Z
--- license: mit tags: - pytorch - diffusers - unconditional-image-generation - diffusion-models-class --- # Model Card for Unit 1 of the [Diffusion Models Class 🧨](https://github.com/huggingface/diffusion-models-class) This model is a diffusion model for unconditional image generation of cute 🦋. ## Usage ```python from diffusers import DDPMPipeline pipeline = DDPMPipeline.from_pretrained('wyb1001/sd-class-butterflies-64') image = pipeline().images[0] image ```
PranavHarshan/merged-NewMes-v15sdf
PranavHarshan
2024-10-13T04:14:55Z
5
0
transformers
[ "transformers", "safetensors", "llama", "text-generation", "conversational", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T04:02:17Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
OverTheCloud/bert-base-uncased-issues-128
OverTheCloud
2024-10-13T04:01:46Z
126
0
transformers
[ "transformers", "safetensors", "bert", "fill-mask", "generated_from_trainer", "base_model:google-bert/bert-base-uncased", "base_model:finetune:google-bert/bert-base-uncased", "license:apache-2.0", "autotrain_compatible", "endpoints_compatible", "region:us" ]
fill-mask
2024-10-13T03:45:18Z
--- library_name: transformers license: apache-2.0 base_model: bert-base-uncased tags: - generated_from_trainer model-index: - name: bert-base-uncased-issues-128 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # bert-base-uncased-issues-128 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2341 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 32 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 16 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 2.1014 | 1.0 | 291 | 1.7049 | | 1.6352 | 2.0 | 582 | 1.5080 | | 1.4965 | 3.0 | 873 | 1.3509 | | 1.3996 | 4.0 | 1164 | 1.3444 | | 1.333 | 5.0 | 1455 | 1.2414 | | 1.2871 | 6.0 | 1746 | 1.3665 | | 1.2358 | 7.0 | 2037 | 1.2885 | | 1.2016 | 8.0 | 2328 | 1.3422 | | 1.1692 | 9.0 | 2619 | 1.2215 | | 1.145 | 10.0 | 2910 | 1.1708 | | 1.1269 | 11.0 | 3201 | 1.1325 | | 1.1127 | 12.0 | 3492 | 1.1719 | | 1.0898 | 13.0 | 3783 | 1.2175 | | 1.0759 | 14.0 | 4074 | 1.2070 | | 1.0764 | 15.0 | 4365 | 1.2166 | | 1.0608 | 16.0 | 4656 | 1.2341 | ### Framework versions - Transformers 4.45.1 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.20.0
Erlanggaa/qwen1.5b
Erlanggaa
2024-10-13T03:52:09Z
85
0
transformers
[ "transformers", "safetensors", "qwen2", "text-generation", "chat", "conversational", "en", "arxiv:2407.10671", "base_model:Qwen/Qwen2.5-1.5B", "base_model:finetune:Qwen/Qwen2.5-1.5B", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T03:38:58Z
--- license: apache-2.0 license_link: https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE language: - en pipeline_tag: text-generation base_model: Qwen/Qwen2.5-1.5B tags: - chat library_name: transformers --- # Qwen2.5-1.5B-Instruct ## Introduction Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2: - Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains. - Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots. - **Long-context Support** up to 128K tokens and can generate up to 8K tokens. - **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more. **This repo contains the instruction-tuned 1.5B Qwen2.5 model**, which has the following features: - Type: Causal Language Models - Training Stage: Pretraining & Post-training - Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings - Number of Parameters: 1.54B - Number of Paramaters (Non-Embedding): 1.31B - Number of Layers: 28 - Number of Attention Heads (GQA): 12 for Q and 2 for KV - Context Length: Full 32,768 tokens and generation 8192 tokens For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/). ## Requirements The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`. With `transformers<4.37.0`, you will encounter the following error: ``` KeyError: 'qwen2' ``` ## Quickstart Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents. ```python from transformers import AutoModelForCausalLM, AutoTokenizer model_name = "Qwen/Qwen2.5-1.5B-Instruct" model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype="auto", device_map="auto" ) tokenizer = AutoTokenizer.from_pretrained(model_name) prompt = "Give me a short introduction to large language model." messages = [ {"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."}, {"role": "user", "content": prompt} ] text = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True ) model_inputs = tokenizer([text], return_tensors="pt").to(model.device) generated_ids = model.generate( **model_inputs, max_new_tokens=512 ) generated_ids = [ output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids) ] response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] ``` ## Evaluation & Performance Detailed evaluation results are reported in this [📑 blog](https://qwenlm.github.io/blog/qwen2.5/). For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html). ## Citation If you find our work helpful, feel free to give us a cite. ``` @misc{qwen2.5, title = {Qwen2.5: A Party of Foundation Models}, url = {https://qwenlm.github.io/blog/qwen2.5/}, author = {Qwen Team}, month = {September}, year = {2024} } @article{qwen2, title={Qwen2 Technical Report}, author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan}, journal={arXiv preprint arXiv:2407.10671}, year={2024} } ```
josedossantos/urf-txtEmenta-legalbertpt
josedossantos
2024-10-13T03:51:19Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T03:47:33Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
josedossantos/urf-txtEmenta-legalbertimbau
josedossantos
2024-10-13T03:47:27Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T03:36:26Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 1024 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
josedossantos/urf-txtEmenta-legalbert
josedossantos
2024-10-13T03:36:16Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T03:32:29Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 5481 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
ZeHeru/sd-class-butterflies-32
ZeHeru
2024-10-13T03:29:41Z
48
0
diffusers
[ "diffusers", "safetensors", "pytorch", "unconditional-image-generation", "diffusion-models-class", "license:mit", "diffusers:DDPMPipeline", "region:us" ]
unconditional-image-generation
2024-10-13T03:27:11Z
--- license: mit tags: - pytorch - diffusers - unconditional-image-generation - diffusion-models-class --- # Model Card for Unit 1 of the [Diffusion Models Class 🧨](https://github.com/huggingface/diffusion-models-class) This model is a diffusion model for unconditional image generation of cute 🦋. ## Usage ```python from diffusers import DDPMPipeline pipeline = DDPMPipeline.from_pretrained('ZeHeru/sd-class-butterflies-32') image = pipeline().images[0] image ```
mav23/Nxcode-CQ-7B-orpo-GGUF
mav23
2024-10-13T03:29:41Z
11
0
null
[ "gguf", "code", "text-generation", "arxiv:2403.07691", "license:other", "endpoints_compatible", "region:us", "conversational" ]
text-generation
2024-10-13T02:45:26Z
--- license_name: tongyi-qianwen-research license_link: https://huggingface.co/Qwen/CodeQwen1.5-7B/blob/main/LICENSE tags: - code pipeline_tag: text-generation license: other --- <a href="https://ntq.com.vn" target="_blank"><img src="https://cdn-uploads.huggingface.co/production/uploads/5ee1b417636bdb3834e2da19/etbfTJuVdAub2evNP_E4g.png" width="200"/></a> ## Introduction Nxcode-CQ-7B-orpo is an [Monolithic Preference Optimization without Reference Model](https://arxiv.org/abs/2403.07691) fine-tune of Qwen/CodeQwen1.5-7B on 100k samples of high-quality ranking data. ## [Evalplus](https://github.com/evalplus/evalplus) | EvalPlus | pass@1 | | --- | --- | | HumanEval | 86.6 | | HumanEval+ | 83.5 | | MBPP(v0.2.0) | 82.3 | | MBPP+(v0.2.0) | 70.4 | We use a simple template to generate the solution for evalplus: ```python "Complete the following Python function:\n{prompt}" ``` [Evalplus Leaderboard](https://evalplus.github.io/leaderboard.html) | Models | HumanEval | HumanEval+| |------ | ------ | ------ | | GPT-4-Turbo (April 2024)| 90.2| 86.6| | GPT-4 (May 2023)| 88.4| 81.17| | GPT-4-Turbo (Nov 2023)| 85.4| 79.3| | CodeQwen1.5-7B-Chat| 83.5| 78.7| | claude-3-opus (Mar 2024)| 82.9| 76.8| | DeepSeek-Coder-33B-instruct| 81.1| 75.0| | WizardCoder-33B-V1.1| 79.9| 73.2| | OpenCodeInterpreter-DS-33B| 79.3| 73.8| | speechless-codellama-34B-v2.0| 77.4| 72| | GPT-3.5-Turbo (Nov 2023)| 76.8| 70.7| | Llama3-70B-instruct| 76.2| 70.7| ## Bigcode Leaderboard [Bigcode Leaderboard](https://huggingface.co/spaces/bigcode/bigcode-models-leaderboard) **09/05/2024** Top 1 average score. Top 2 winrate. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5ee1b417636bdb3834e2da19/OQonD6a7aNjnN9SsTkFp-.png) ## Quickstart Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents. You should upgrade the transformers if you receive an error when loading the tokenizer ```python from transformers import AutoModelForCausalLM, AutoTokenizer device = "cuda" # the device to load the model onto model = AutoModelForCausalLM.from_pretrained( "NTQAI/Nxcode-CQ-7B-orpo", torch_dtype="auto", device_map="auto" ) tokenizer = AutoTokenizer.from_pretrained("NTQAI/Nxcode-CQ-7B-orpo") prompt = """Complete the following Python function: from typing import List def has_close_elements(numbers: List[float], threshold: float) -> bool: """ Check if in given list of numbers, are any two numbers closer to each other than given threshold. >>> has_close_elements([1.0, 2.0, 3.0], 0.5) False >>> has_close_elements([1.0, 2.8, 3.0, 4.0, 5.0, 2.0], 0.3) True """ """ messages = [ {"role": "user", "content": prompt} ] inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device) outputs = model.generate(inputs, max_new_tokens=512, do_sample=False, top_k=50, top_p=0.95, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id) res = tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True) ``` ### Contact information For personal communication related to this project, please contact Nha Nguyen Van ([email protected]).
gayfortay13/Algo
gayfortay13
2024-10-13T03:27:37Z
50
0
diffusers
[ "diffusers", "text-to-image", "lora", "template:diffusion-lora", "base_model:black-forest-labs/FLUX.1-dev", "base_model:adapter:black-forest-labs/FLUX.1-dev", "region:us" ]
text-to-image
2024-10-13T03:25:07Z
--- tags: - text-to-image - lora - diffusers - template:diffusion-lora widget: - text: '-' output: url: images/0822319e-5724-4966-acd9-c20ea5e7662a.jpeg base_model: black-forest-labs/FLUX.1-dev instance_prompt: Taylor Swift --- # TaylorSwizzle <Gallery /> ## Trigger words You should use `Taylor Swift` to trigger the image generation. ## Download model Weights for this model are available in Safetensors format. [Download](/gayfortay13/Algo/tree/main) them in the Files & versions tab.
FuturisticVibes/Mistral-Nemo-12B-ArliAI-RPMax-v1.2-6.0bpw-h8-exl2
FuturisticVibes
2024-10-13T03:07:33Z
8
1
null
[ "safetensors", "mistral", "license:apache-2.0", "6-bit", "exl2", "region:us" ]
null
2024-10-13T03:04:06Z
--- license: apache-2.0 --- I have no idea what I’m doing… if this causes the apocalypse someone please let me know. Mistral-Nemo-12B-ArliAI-RPMax-v1.2 6.0bpw h8 EXL2 Includes [measurement.json](https://huggingface.co/FuturisticVibes/Mistral-Nemo-12B-ArliAI-RPMax-v1.2-6.0bpw-h8-exl2/tree/measurement) file for further quantization Original Model: https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2/ # Original Model Card # ArliAI-RPMax-12B-v1.2 ===================================== ## UPDATE: For those getting gibberish results, it was merged wrongly to base after LORA training. Reuploaded all the files so it should work properly now. ## RPMax Series Overview | [2B](https://huggingface.co/ArliAI/Gemma-2-2B-ArliAI-RPMax-v1.1) | [3.8B](https://huggingface.co/ArliAI/Phi-3.5-mini-3.8B-ArliAI-RPMax-v1.1) | [8B](https://huggingface.co/ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.2) | [9B](https://huggingface.co/ArliAI/Gemma-2-9B-ArliAI-RPMax-v1.1) | [12B](https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2) | [20B](https://huggingface.co/ArliAI/InternLM2_5-20B-ArliAI-RPMax-v1.1) | [22B](https://huggingface.co/ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1) | [70B](https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1) | RPMax is a series of models that are trained on a diverse set of curated creative writing and RP datasets with a focus on variety and deduplication. This model is designed to be highly creative and non-repetitive by making sure no two entries in the dataset have repeated characters or situations, which makes sure the model does not latch on to a certain personality and be capable of understanding and acting appropriately to any characters or situations. Early tests by users mentioned that these models does not feel like any other RP models, having a different style and generally doesn't feel in-bred. You can access the model at https://arliai.com and ask questions at https://www.reddit.com/r/ArliAI/ We also have a models ranking page at https://www.arliai.com/models-ranking Ask questions in our new Discord Server! https://discord.com/invite/t75KbPgwhk ## Model Description ArliAI-RPMax-12B-v1.2 is a variant based on Mistral Nemo 12B Instruct 2407. This is arguably the most successful RPMax model due to how Mistral is already very uncensored in the first place. v1.2 update completely removes non-creative/RP examples in the dataset and is also an incremental improvement of the RPMax dataset which dedups the dataset even more and better filtering to cutout irrelevant description text that came from card sharing sites. ### Specs * **Context Length**: 128K * **Parameters**: 12B ### Training Details * **Sequence Length**: 8192 * **Training Duration**: Approximately 2 days on 2x3090Ti * **Epochs**: 1 epoch training for minimized repetition sickness * **LORA**: 64-rank 128-alpha, resulting in ~2% trainable weights * **Learning Rate**: 0.00001 * **Gradient accumulation**: Very low 32 for better learning. ## Quantization The model is available in quantized formats: * **FP16**: https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2 * **GGUF**: https://huggingface.co/ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2-GGUF ## Suggested Prompt Format Mistral Instruct Prompt Format
yukihara00/speecht5_finetuned_voxpopuli_nl
yukihara00
2024-10-13T02:59:45Z
76
0
transformers
[ "transformers", "tensorboard", "safetensors", "speecht5", "text-to-audio", "generated_from_trainer", "dataset:voxpopuli", "base_model:microsoft/speecht5_tts", "base_model:finetune:microsoft/speecht5_tts", "license:mit", "endpoints_compatible", "region:us" ]
text-to-audio
2024-10-13T01:22:47Z
--- library_name: transformers license: mit base_model: microsoft/speecht5_tts tags: - generated_from_trainer datasets: - voxpopuli model-index: - name: speecht5_finetuned_voxpopuli_nl results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # speecht5_finetuned_voxpopuli_nl This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the voxpopuli dataset. It achieves the following results on the evaluation set: - Loss: 0.4595 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 4 - eval_batch_size: 2 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - training_steps: 4000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-------:|:----:|:---------------:| | 0.5193 | 4.3034 | 1000 | 0.4794 | | 0.4967 | 8.6068 | 2000 | 0.4669 | | 0.4893 | 12.9102 | 3000 | 0.4616 | | 0.4918 | 17.2136 | 4000 | 0.4595 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.4.1 - Datasets 3.0.1 - Tokenizers 0.19.1
murugeshmarvel/QAD_LLM_iter3_set2_llama3.1_best
murugeshmarvel
2024-10-13T02:54:58Z
7
0
transformers
[ "transformers", "safetensors", "llama", "text-generation", "conversational", "arxiv:1910.09700", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T02:44:28Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
mav23/Llama3-Chinese-8B-Instruct-GGUF
mav23
2024-10-13T02:40:33Z
65
0
null
[ "gguf", "llama3", "chinese", "license:apache-2.0", "endpoints_compatible", "region:us", "conversational" ]
null
2024-10-13T01:50:14Z
--- license: apache-2.0 tags: - llama3 - chinese --- # Llama3-Chinese-8B-Instruct Llama3-Chinese-8B-Instruct基于Llama3-8B中文微调对话模型,由Llama中文社区和AtomEcho(原子回声)联合研发,我们会持续提供更新的模型参数,模型训练过程见 [https://llama.family](https://llama.family)。 模型的部署、训练、微调等方法详见Llama中文社区GitHub仓库:[https://github.com/LlamaFamily/Llama-Chinese](https://github.com/LlamaFamily/Llama-Chinese) ## 如何使用 ``` import transformers import torch model_id = "FlagAlpha/Llama3-Chinese-8B-Instruct" pipeline = transformers.pipeline( "text-generation", model=model_id, model_kwargs={"torch_dtype": torch.float16}, device="cuda", ) messages = [{"role": "system", "content": ""}] messages.append( {"role": "user", "content": "介绍一下机器学习"} ) prompt = pipeline.tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True ) terminators = [ pipeline.tokenizer.eos_token_id, pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>") ] outputs = pipeline( prompt, max_new_tokens=512, eos_token_id=terminators, do_sample=True, temperature=0.6, top_p=0.9 ) content = outputs[0]["generated_text"][len(prompt):] print(content) ```
LyubomirT/gpt2-student-suggester
LyubomirT
2024-10-13T02:40:19Z
122
0
transformers
[ "transformers", "safetensors", "gpt2", "text-generation", "gpt-2", "education", "student-advisor", "suggestions", "casual-lm", "en", "base_model:openai-community/gpt2", "base_model:finetune:openai-community/gpt2", "license:mit", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T02:23:51Z
--- license: mit language: - en base_model: - openai-community/gpt2 pipeline_tag: text-generation library_name: transformers tags: - gpt-2 - text-generation - education - student-advisor - suggestions - casual-lm --- # GPT2 Student Advisor ## Model Overview Meet the **GPT2 Student Advisor**, your personal AI counselor who’s been fine-tuned on student data to dole out academic advice. This model is based on GPT-2 and has been trained to generate personalized suggestions for students. Think of it as a non-judgmental guide to tell you (or your students) that maybe skipping classes and not studying isn’t the best strategy. It looks at stuff like study hours, attendance, sleep patterns, and other school-related activities to give advice, whether it’s to buckle down, sleep more, or, in some cases, keep up the good work. ### Model Architecture - **Base model**: GPT-2 (because GPT-3 was busy) - **Fine-tuned**: Yep! Specifically for generating suggestions based on student profiles. ## What’s This For? This isn’t just some random text generator. The **GPT2 Student Advisor** can actually be used to give academic guidance, helping students (or teachers) with actionable advice. It’s especially handy for: - **Automated Student Advising**: Use it as a chatbot to gently nudge students in the right direction. - **Educational Platforms**: Got an education app? Plug this in for a tailored learning experience. - **Counselor’s Best Friend**: Helping counselors generate quick suggestions (but seriously, humans are still important). ## Training Data The model was trained on the **Student Performance Factors** dataset, which has data on things like: - Study Hours - Attendance - Parental Involvement (aka “Do your parents care about your grades?”) - Sleep Hours (yes, sleep matters) - Internet Access (because no internet is the modern tragedy) - Plus a bunch of other factors that make students tick. Using this data, the model learned to generate suggestions like "Maybe study more than 5 hours a week" or "Get some sleep." All the good stuff. ### Sample Input: ``` Student Profile: - Hours Studied per week: 5 - Attendance: 60% - Parental Involvement: Low - Access to Resources: Medium - Extracurricular Activities: None - Sleep Hours per night: 6 - Previous Scores: 70 - Motivation Level: Low - Internet Access: None - Tutoring Sessions per month: 0 - Family Income: Low - Teacher Quality: Medium - School Type: Public - Peer Influence: Negative - Physical Activity per week: 1 hour - Learning Disabilities: Yes - Parental Education Level: High School - Distance from Home: Far - Gender: Male ``` ### Sample Output: ``` Suggestions: - Consider increasing your study hours. - Improve your class attendance. - Seek more support from your parents. - Ensure you get enough sleep each night. - Find ways to boost your motivation. - Find ways to access the internet for study resources. - Consider seeking help for your learning disabilities. - Engage in more physical activities for better health. ``` ## How We Trained It - **Batch size**: 8 (because 16 was too easy) - **Epochs**: 3 (the magic number) - **Learning rate**: 5e-5 - **Optimizer**: Good ol’ AdamW - **Evaluation**: After each epoch, because we like checking in on progress. - **Mixed precision**: Enabled, because who doesn't like faster training? ### Training Environment - **Hardware**: Powered by an NVIDIA GPU (because anything else just wouldn't cut it). - **Software**: `transformers` library from Hugging Face, running on PyTorch. ## Performance The model was evaluated using: - **Loss**: Yeah, loss matters here too. We minimized it using causal language modeling. - **Best Model Selection**: After every epoch, we kept the one that made the least mistakes. Low loss = good. ## Things to Keep in Mind - **It's Specialized**: This model was trained on student data. Don’t expect it to give life advice outside the academic sphere—though it might be fun to ask! - **Be Specific**: Keep the student profile format consistent for the best results. Random input = random output. ## Ethical Stuff Let’s not pretend this model is a replacement for real counseling. It’s trained on a dataset that makes a lot of assumptions about student behavior and performance. Real life is, of course, more complex. Please use this model as an *assistant*, not a replacement for human guidance. ## How to Use It ```python from transformers import GPT2LMHeadModel, GPT2Tokenizer # Load the fine-tuned model and tokenizer model = GPT2LMHeadModel.from_pretrained("LyubomirT/gpt2-student-suggester") tokenizer = GPT2Tokenizer.from_pretrained("LyubomirT/gpt2-student-suggester") # Define a student profile student_profile = """ Student Profile: - Hours Studied per week: 5 - Attendance: 60% - Parental Involvement: Low - Sleep Hours per night: 6 - Motivation Level: Low """ # Tokenize the input and generate suggestions inputs = tokenizer.encode(student_profile, return_tensors="pt") outputs = model.generate(inputs, max_length=600, num_beams=5, early_stopping=True) suggestions = tokenizer.decode(outputs[0], skip_special_tokens=True) print("Generated Suggestions:") print(suggestions) ``` ## License This model is released under the MIT license, because we believe in sharing the love. Check out [Hugging Face's Model Licensing guidelines](https://huggingface.co/docs/hub/model-repositories#license-a-model) for more info.
ammonbro/up_down_sp_model
ammonbro
2024-10-13T02:37:54Z
180
0
transformers
[ "transformers", "safetensors", "gpt2", "text-generation", "generated_from_trainer", "base_model:distilbert/distilgpt2", "base_model:finetune:distilbert/distilgpt2", "license:apache-2.0", "autotrain_compatible", "text-generation-inference", "endpoints_compatible", "region:us" ]
text-generation
2024-10-13T01:12:30Z
--- library_name: transformers license: apache-2.0 base_model: distilgpt2 tags: - generated_from_trainer model-index: - name: up_down_sp_model results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # up_down_sp_model This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.7062 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3.0 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | 0.6954 | 1.0 | 1429 | 0.7057 | | 0.691 | 2.0 | 2858 | 0.7039 | | 0.6887 | 3.0 | 4287 | 0.7062 | ### Framework versions - Transformers 4.45.2 - Pytorch 2.4.1+cu121 - Datasets 3.0.1 - Tokenizers 0.20.1
zhangtaolab/plant-dnagpt-6mer-promoter_strength_leaf
zhangtaolab
2024-10-13T02:35:07Z
5
0
null
[ "safetensors", "gpt2", "DNA", "biology", "genomics", "license:cc-by-nc-sa-4.0", "region:us" ]
null
2024-10-13T02:34:08Z
--- license: cc-by-nc-sa-4.0 widget: - text: AGTCCAGTGGACGACCAGCCACGGCTCCGGTCTGTAGAACCATCGCGGAAACGGCTCGCAAAACTCTAAACAGCGCAAACGATGCGCGCGCCGAAGCAACCCGGCTCTACTTATAAAAACGTCCAACGGTGAGCACCGAGCAGCTACTACTCGTACTCCCCCCACCGATC tags: - DNA - biology - genomics --- # Plant foundation DNA large language models The plant DNA large language models (LLMs) contain a series of foundation models based on different model architectures, which are pre-trained on various plant reference genomes. All the models have a comparable model size between 90 MB and 150 MB, BPE tokenizer is used for tokenization and 8000 tokens are included in the vocabulary. **Developed by:** zhangtaolab ### Model Sources - **Repository:** [Plant DNA LLMs](https://github.com/zhangtaolab/plant_DNA_LLMs) - **Manuscript:** [Versatile applications of foundation DNA language models in plant genomes]() ### Architecture The model is trained based on the OpenAI GPT-2 model with modified tokenizer specific for DNA sequence. This model is fine-tuned for predicting promoter strength in tobacco leaves system. ### How to use Install the runtime library first: ```bash pip install transformers ``` Here is a simple code for inference: ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline model_name = 'plant-dnagpt-6mer-promoter_strength_leaf' # load model and tokenizer model = AutoModelForSequenceClassification.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) # inference sequences = ['TACTCTAATCGTATCAGCTGCACTTGCGTACAGGCTACCGGCGTCCTCAGCCACGTAAGAAAAGGCCCAATAAAGGCCCAACTACAACCAGCGGATATATATACTGGAGCCTGGCGAGATCACCCTAACCCCTCACACTCCCATCCAGCCGCCACCAGGTGCAGAGTGTT', 'ATTTCAAAACTAGTTTTCTATAAACGAAAACTTATATTTATTCCGCTTGTTCCGTTTGATCTGCTGATTCGACACCGTTTTAACGTATTTTAAGTAAGTATCAGAAATATTAATGTGAAGATAAAAGAAAATAGAGTAAATGTAAAGGAAAATGCATAAGATTTTGTTGA'] pipe = pipeline('text-classification', model=model, tokenizer=tokenizer, trust_remote_code=True, function_to_apply="none") results = pipe(sequences) print(results) ``` ### Training data We use GPT2ForSequenceClassification to fine-tune the model. Detailed training procedure can be found in our manuscript. #### Hardware Model was trained on a NVIDIA GTX1080Ti GPU (11 GB).
bhavikardeshna/xlm-roberta-base-vietnamese
bhavikardeshna
2024-10-13T02:32:13Z
142
1
transformers
[ "transformers", "pytorch", "safetensors", "xlm-roberta", "question-answering", "arxiv:2112.09866", "endpoints_compatible", "region:us" ]
question-answering
2022-03-02T23:29:05Z
# BibTeX entry and citation info ``` @misc{pandya2021cascading, title={Cascading Adaptors to Leverage English Data to Improve Performance of Question Answering for Low-Resource Languages}, author={Hariom A. Pandya and Bhavik Ardeshna and Dr. Brijesh S. Bhatt}, year={2021}, eprint={2112.09866}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```
blanchon/PixDiet-pixtral-nutrition-v2
blanchon
2024-10-13T02:31:22Z
13
1
transformers
[ "transformers", "tensorboard", "safetensors", "llava", "image-text-to-text", "arxiv:1910.09700", "endpoints_compatible", "region:us" ]
image-text-to-text
2024-10-12T21:06:41Z
--- library_name: transformers tags: [] --- # Model Card for Model ID <!-- Provide a quick summary of what the model is/does. --> ## Model Details ### Model Description <!-- Provide a longer summary of what this model is. --> This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated. - **Developed by:** [More Information Needed] - **Funded by [optional]:** [More Information Needed] - **Shared by [optional]:** [More Information Needed] - **Model type:** [More Information Needed] - **Language(s) (NLP):** [More Information Needed] - **License:** [More Information Needed] - **Finetuned from model [optional]:** [More Information Needed] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [More Information Needed] - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] ## Uses <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> ### Direct Use <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> [More Information Needed] ### Downstream Use [optional] <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app --> [More Information Needed] ### Out-of-Scope Use <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> [More Information Needed] ## Bias, Risks, and Limitations <!-- This section is meant to convey both technical and sociotechnical limitations. --> [More Information Needed] ### Recommendations <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. --> Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. ## How to Get Started with the Model Use the code below to get started with the model. [More Information Needed] ## Training Details ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> [More Information Needed] ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. --> [More Information Needed] #### Factors <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. --> [More Information Needed] #### Metrics <!-- These are the evaluation metrics being used, ideally with a description of why. --> [More Information Needed] ### Results [More Information Needed] #### Summary ## Model Examination [optional] <!-- Relevant interpretability work for the model goes here --> [More Information Needed] ## Environmental Impact <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). - **Hardware Type:** [More Information Needed] - **Hours used:** [More Information Needed] - **Cloud Provider:** [More Information Needed] - **Compute Region:** [More Information Needed] - **Carbon Emitted:** [More Information Needed] ## Technical Specifications [optional] ### Model Architecture and Objective [More Information Needed] ### Compute Infrastructure [More Information Needed] #### Hardware [More Information Needed] #### Software [More Information Needed] ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> **BibTeX:** [More Information Needed] **APA:** [More Information Needed] ## Glossary [optional] <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. --> [More Information Needed] ## More Information [optional] [More Information Needed] ## Model Card Authors [optional] [More Information Needed] ## Model Card Contact [More Information Needed]
zhangtaolab/plant-dnagpt-6mer-H3K27ac
zhangtaolab
2024-10-13T02:28:22Z
5
0
null
[ "safetensors", "gpt2", "DNA", "biology", "genomics", "license:cc-by-nc-sa-4.0", "region:us" ]
null
2024-10-13T02:27:27Z
--- license: cc-by-nc-sa-4.0 widget: - text: GCGACTCCGCCGCCCCGATCTCCCCGTCGTCCTACAGTGCTCTCCACATCGTAGGCGACCTGGTTGGACTCCTCGACGCCTTGTCCCTACCGCAGGTGTTTGTGGTGGGACAAGGCTGGGGAGCCCTGCTGGCGTGGAACCTCTGCATGTTCCGCCCCGAGCGGGTGCGCGCGCTGGTCAACATGAGCGTCGCCTTCATGCCGCGCAACCCCTCCGTGAAGCCACTTGAGTTGTTTCGGCGGCTCTACGGCGACGGATACTACCTCCTCCGGCTGCAGGAAC tags: - DNA - biology - genomics --- # Plant foundation DNA large language models The plant DNA large language models (LLMs) contain a series of foundation models based on different model architectures, which are pre-trained on various plant reference genomes. All the models have a comparable model size between 90 MB and 150 MB, BPE tokenizer is used for tokenization and 8000 tokens are included in the vocabulary. **Developed by:** zhangtaolab ### Model Sources - **Repository:** [Plant DNA LLMs](https://github.com/zhangtaolab/plant_DNA_LLMs) - **Manuscript:** [Versatile applications of foundation DNA language models in plant genomes]() ### Architecture The model is trained based on the OpenAI GPT-2 model with modified tokenizer specific for DNA sequence. This model is fine-tuned for predicting H3K4me3 histone modification. ### How to use Install the runtime library first: ```bash pip install transformers ``` Here is a simple code for inference: ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline model_name = 'plant-dnagpt-6mer-H3K4me3' # load model and tokenizer model = AutoModelForSequenceClassification.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) # inference sequences = ['TTCATCTCGTCCGACGCTTCAACCCGCACCGATCCTGCGCCACCCCTTCGCCGGCGGCTTCTCCCCTCCTCTTCCTCCGCCGCTGCATCGCCGTCCCAGGAACTTGGACACGTCGCCTCTCGCCGGCGACCATGTACCGCGCCCTCCGCTCTCTCAAGGTTTCCCCGTCTGCACCCCCCCAACCTTCTACGACGTGTGGCGTTGCGTGTCTCGATCCATTTGGGATGAATGCGCTGGAGTGTTAGA', 'ATCAATATTCCCAACAGGTTTTGAAGCAATGGATGAAACATCATCCTTCACGGAACTGGATTATGGGATTCGCCGGCTGGACCACGCTGTTGGGAATGTGCCGGAGCTGGGTCCTGTAGTGGATTACATCAAGGCGTTTACGGGGTTTCATGAATTTGCGGAGTTTACAGCT'] pipe = pipeline('text-classification', model=model, tokenizer=tokenizer, trust_remote_code=True, top_k=None) results = pipe(sequences) print(results) ``` ### Training data We use GPT2ForSequenceClassification to fine-tune the model. Detailed training procedure can be found in our manuscript. #### Hardware Model was trained on a NVIDIA GTX1080Ti GPU (11 GB).
josedossantos/tree-txtInteiroTeorLimpo-legalbert
josedossantos
2024-10-13T02:27:22Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T02:26:50Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, mean pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 2164 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
zhangtaolab/plant-dnagpt-6mer-H3K4me3
zhangtaolab
2024-10-13T02:24:59Z
8
0
null
[ "safetensors", "gpt2", "DNA", "biology", "genomics", "license:cc-by-nc-sa-4.0", "region:us" ]
null
2024-10-13T02:24:01Z
--- license: cc-by-nc-sa-4.0 widget: - text: GCGACTCCGCCGCCCCGATCTCCCCGTCGTCCTACAGTGCTCTCCACATCGTAGGCGACCTGGTTGGACTCCTCGACGCCTTGTCCCTACCGCAGGTGTTTGTGGTGGGACAAGGCTGGGGAGCCCTGCTGGCGTGGAACCTCTGCATGTTCCGCCCCGAGCGGGTGCGCGCGCTGGTCAACATGAGCGTCGCCTTCATGCCGCGCAACCCCTCCGTGAAGCCACTTGAGTTGTTTCGGCGGCTCTACGGCGACGGATACTACCTCCTCCGGCTGCAGGAAC tags: - DNA - biology - genomics --- # Plant foundation DNA large language models The plant DNA large language models (LLMs) contain a series of foundation models based on different model architectures, which are pre-trained on various plant reference genomes. All the models have a comparable model size between 90 MB and 150 MB, BPE tokenizer is used for tokenization and 8000 tokens are included in the vocabulary. **Developed by:** zhangtaolab ### Model Sources - **Repository:** [Plant DNA LLMs](https://github.com/zhangtaolab/plant_DNA_LLMs) - **Manuscript:** [Versatile applications of foundation DNA language models in plant genomes]() ### Architecture The model is trained based on the OpenAI GPT-2 model with modified tokenizer specific for DNA sequence. This model is fine-tuned for predicting H3K4me3 histone modification. ### How to use Install the runtime library first: ```bash pip install transformers ``` Here is a simple code for inference: ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline model_name = 'plant-dnagpt-6mer-H3K4me3' # load model and tokenizer model = AutoModelForSequenceClassification.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) # inference sequences = ['TTCATCTCGTCCGACGCTTCAACCCGCACCGATCCTGCGCCACCCCTTCGCCGGCGGCTTCTCCCCTCCTCTTCCTCCGCCGCTGCATCGCCGTCCCAGGAACTTGGACACGTCGCCTCTCGCCGGCGACCATGTACCGCGCCCTCCGCTCTCTCAAGGTTTCCCCGTCTGCACCCCCCCAACCTTCTACGACGTGTGGCGTTGCGTGTCTCGATCCATTTGGGATGAATGCGCTGGAGTGTTAGA', 'ATCAATATTCCCAACAGGTTTTGAAGCAATGGATGAAACATCATCCTTCACGGAACTGGATTATGGGATTCGCCGGCTGGACCACGCTGTTGGGAATGTGCCGGAGCTGGGTCCTGTAGTGGATTACATCAAGGCGTTTACGGGGTTTCATGAATTTGCGGAGTTTACAGCT'] pipe = pipeline('text-classification', model=model, tokenizer=tokenizer, trust_remote_code=True, top_k=None) results = pipe(sequences) print(results) ``` ### Training data We use GPT2ForSequenceClassification to fine-tune the model. Detailed training procedure can be found in our manuscript. #### Hardware Model was trained on a NVIDIA GTX1080Ti GPU (11 GB).
zhangtaolab/plant-dnagpt-singlebase-open_chromatin
zhangtaolab
2024-10-13T02:17:05Z
5
0
null
[ "safetensors", "gpt2", "DNA", "biology", "genomics", "license:cc-by-nc-sa-4.0", "region:us" ]
null
2024-10-13T02:16:04Z
--- license: cc-by-nc-sa-4.0 widget: - text: GCAAGGTGGGTTTGGTCTCTGTCTGGTACGTAGAGGAGAAAGAGACGAAGGGGATAGGAAGAGAGATGATGGTCAAAATATGTATCTAAGTAGATGTATAGGTATTTGACAAAATATAGATATTTATCTAATTAATAGTTCATGTGTCTGGTAAAGTGTAC tags: - DNA - biology - genomics --- # Plant foundation DNA large language models The plant DNA large language models (LLMs) contain a series of foundation models based on different model architectures, which are pre-trained on various plant reference genomes. All the models have a comparable model size between 90 MB and 150 MB, BPE tokenizer is used for tokenization and 8000 tokens are included in the vocabulary. **Developed by:** zhangtaolab ### Model Sources - **Repository:** [Plant DNA LLMs](https://github.com/zhangtaolab/plant_DNA_LLMs) - **Manuscript:** [Versatile applications of foundation DNA large language models in plant genomes]() ### Architecture The model is trained based on the OpenAI GPT-2 model with modified tokenizer specific for DNA sequence. This model is fine-tuned for predicting open chromatin. ### How to use Install the runtime library first: ```bash pip install transformers ``` Here is a simple code for inference: ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline model_name = 'plant-dnagpt-singlebase-open_chromatin' # load model and tokenizer model = AutoModelForSequenceClassification.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) # inference sequences = ['TTTTGATTCAGTGATTTTCGTCCTTTACAAAAGCTAATCCTTTTGGCCGCTTGACATAGATGATGCAGATCTTATCTGAATATCATTCCAGGTGCGTCGCGAGGGAATTGCTGTCGCGAATCGATCGATAAGAGACGGCTGGGTACGGGGTGGGTATGGATATGAACTTTTGCTTCC', 'GATGCTACTGCTAGCTAATCAGTAATCACCAATGCATAAACACAACACATGCCTTCGTTCCAAAGTTTTCATTCCTCGTCATAGACTTAAAGAAGGGGCAACAAGTTCTCTACGAGTCTTCTGGACTGGACTGGCTACCCCCTCGGCCCATTCTGGCCCAGTTGCGGGCGGCCTTTCATTTAATAAATATTTCTAATAGATATAAATTATTTTATCTAATATTATTAATTTTTTTCTTATAAAACATATAAT'] pipe = pipeline('text-classification', model=model, tokenizer=tokenizer, trust_remote_code=True, top_k=None) results = pipe(sequences) print(results) ``` ### Training data We use GPT2ForSequenceClassification to fine-tune the model. Detailed training procedure can be found in our manuscript. #### Hardware Model was trained on a NVIDIA GTX1080Ti GPU (11 GB).
zhangtaolab/plant-dnagpt-singlebase-lncRNAs
zhangtaolab
2024-10-13T02:15:29Z
5
0
null
[ "safetensors", "gpt2", "DNA", "biology", "genomics", "license:cc-by-nc-sa-4.0", "region:us" ]
null
2024-10-13T02:14:32Z
--- license: cc-by-nc-sa-4.0 widget: - text: ATGCTTTGTGCTGGCATGCCATGTCATGTTGCATCAGCATTTTCTTTATATTTTCTTTCTGATCTTTTCTGTGCTTCAAAACCTCATTCGTCTGTTTCCTTCTTTCCTACCAGTTATCCACAGACACACCCTATTAGAGTACTCCATGCTTGTTTATTTCTTTTGTCAAATAGAAGGGTCTTTTCTCCTCGCTTTAGTAGGGAATGTTGTCTTCCTCATTTGGGAAAAAAAAATTGTTCCTGCAGTTATGCCAGTCATGGGCTCTTTTTGATTGGTTGCATTGATATATTGTCTACCCCGTTTTCTGTAGGAATGATACATATTCCTGATCCTGAGCCTATTTGA tags: - DNA - biology - genomics --- # Plant foundation DNA large language models The plant DNA large language models (LLMs) contain a series of foundation models based on different model architectures, which are pre-trained on various plant reference genomes. All the models have a comparable model size between 90 MB and 150 MB, BPE tokenizer is used for tokenization and 8000 tokens are included in the vocabulary. **Developed by:** zhangtaolab ### Model Sources - **Repository:** [Plant DNA LLMs](https://github.com/zhangtaolab/plant_DNA_LLMs) - **Manuscript:** [Versatile applications of foundation DNA large language models in plant genomes]() ### Architecture The model is trained based on the OpenAI GPT-2 model with modified tokenizer specific for DNA sequence. This model is fine-tuned for predicting lncRNAs. ### How to use Install the runtime library first: ```bash pip install transformers ``` Here is a simple code for inference: ```python from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline model_name = 'plant-dnagpt-singlebase-lncRNAs' # load model and tokenizer model = AutoModelForSequenceClassification.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained(f'zhangtaolab/{model_name}', trust_remote_code=True) # inference sequences = ['ATGGAACTCATGAAGACGTTAGATCTTCACAAGAGGATATTTTCCGAATTTAGTGATGAACAATCAAGAGTGTCATACACTGCAAAAATCTATCAAGAACAAATAAAAGCGGCAAAAGGGAGGTTGCCTGATAGTAGTGTAAAGCAATTAGGTGTCTGGCAACTTCATGTTTTCCTCAAAAGATGTGAAAAAGCACCCAACCAGGACAATACGACATCAGGAATTCTGTAA', 'ATGGCTGATGAAGCTCAGGAGAAGGCTGAACATGATCGCATTTTCAAGCGCTTCGACTTGAACGGAGACGGCAGGATCTCTGCCGCAGAGCTAGGTGACTGCTTGAAGACCCTTGGTTCAGTCACCCCGGATGAGATCCAGCGTATGATGGCAGAGATTGATACTGATGGTGATGGATACATATCATATGAAGAATTCACAGATTTTGCCATGGCCAACCGTGGCCTAATGAAGGATGTGGCCAAGATATTCTAA'] pipe = pipeline('text-classification', model=model, tokenizer=tokenizer, trust_remote_code=True, top_k=None) results = pipe(sequences) print(results) ``` ### Training data We use GPT2ForSequenceClassification to fine-tune the model. Detailed training procedure can be found in our manuscript. #### Hardware Model was trained on a NVIDIA GTX1080Ti GPU (11 GB).
josedossantos/tree-txtInteiroTeorLimpo-labse
josedossantos
2024-10-13T01:59:38Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "feature-extraction", "sentence-similarity", "transformers", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T01:58:07Z
--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # {MODEL_NAME} This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. <!--- Describe your model here --> ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('{MODEL_NAME}') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch def cls_pooling(model_output, attention_mask): return model_output[0][:,0] # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}') model = AutoModel.from_pretrained('{MODEL_NAME}') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, cls pooling. sentence_embeddings = cls_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results <!--- Describe how your model was evaluated --> For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name={MODEL_NAME}) ## Training The model was trained with the parameters: **DataLoader**: `torch.utils.data.dataloader.DataLoader` of length 2164 with parameters: ``` {'batch_size': 2, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'} ``` **Loss**: `sentence_transformers.losses.ContrastiveLoss.ContrastiveLoss` with parameters: ``` {'distance_metric': 'SiameseDistanceMetric.COSINE_DISTANCE', 'margin': 0.5, 'size_average': True} ``` Parameters of the fit()-Method: ``` { "epochs": 1, "evaluation_steps": 0, "evaluator": "NoneType", "max_grad_norm": 1, "optimizer_class": "<class 'torch.optim.adamw.AdamW'>", "optimizer_params": { "lr": 2e-05 }, "scheduler": "WarmupLinear", "steps_per_epoch": null, "warmup_steps": 10000, "weight_decay": 0.01 } ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Citing & Authors <!--- Describe where people can find more information -->
FM-1976/gemma-2-2b-it-Q5_K_M-GGUF
FM-1976
2024-10-13T01:56:54Z
12
1
transformers
[ "transformers", "gguf", "conversational", "llama-cpp", "gguf-my-repo", "text-generation", "base_model:google/gemma-2-2b-it", "base_model:quantized:google/gemma-2-2b-it", "license:gemma", "endpoints_compatible", "region:us" ]
text-generation
2024-10-12T08:40:27Z
--- base_model: google/gemma-2-2b-it library_name: transformers license: gemma pipeline_tag: text-generation tags: - conversational - llama-cpp - gguf-my-repo extra_gated_heading: Access Gemma on Hugging Face extra_gated_prompt: To access Gemma on Hugging Face, you’re required to review and agree to Google’s usage license. To do this, please ensure you’re logged in to Hugging Face and click below. Requests are processed immediately. extra_gated_button_content: Acknowledge license --- <img src='https://github.com/fabiomatricardi/Gemma2-2b-it-chatbot/raw/main/images/gemma2-2b-myGGUF.png' width=900> <br><br><br> # FM-1976/gemma-2-2b-it-Q5_K_M-GGUF This model was converted to GGUF format from [`google/gemma-2-2b-it`](https://huggingface.co/google/gemma-2-2b-it) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space. Refer to the [original model card](https://huggingface.co/google/gemma-2-2b-it) for more details on the model. ## Description Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. They are text-to-text, decoder-only large language models, available in English, with open weights for both pre-trained variants and instruction-tuned variants. Gemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as a laptop, desktop or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone. ## Model Details context window = 8192 SYSTEM MESSAGE NOT SUPPORTED ```bash architecture str = gemma2 type str = model name str = Gemma 2 2b It finetune str = it basename str = gemma-2 size_label str = 2B license str = gemma count u32 = 1 model.0.name str = Gemma 2 2b organization str = Google format = GGUF V3 (latest) arch = gemma2 vocab type = SPM n_vocab = 256000 n_merges = 0 vocab_only = 0 n_ctx_train = 8192 n_embd = 2304 n_layer = 26 n_head = 8 n_head_kv = 4 model type = 2B model ftype = Q5_K - Medium model params = 2.61 B model size = 1.79 GiB (5.87 BPW) general.name = Gemma 2 2b It BOS token = 2 '<bos>' EOS token = 1 '<eos>' UNK token = 3 '<unk>' PAD token = 0 '<pad>' LF token = 227 '<0x0A>' EOT token = 107 '<end_of_turn>' EOG token = 1 '<eos>' EOG token = 107 '<end_of_turn>' >>> System role not supported Available chat formats from metadata: chat_template.default Using gguf chat template: {{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + ' ' + message['content'] | trim + '<end_of_turn> ' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model '}}{% endif %} Using chat eos_token: <eos> Using chat bos_token: <bos> ``` ### Prompt Format ```pthon <bos><start_of_turn>user {prompt}<end_of_turn> <start_of_turn>model <end_of_turn> ``` ## Chat Template The instruction-tuned models use a chat template that must be adhered to for conversational use. The easiest way to apply it is using the tokenizer's built-in chat template, as shown in the following snippet. ```python messages = [ {"role": "user", "content": "Write me a poem about Machine Learning."}, ] ``` ## Use with llama-cpp-python Install llama.cpp through brew (works on Mac and Linux) ```bash pip install llama-cpp-python ``` ### Download locally the GGUF file ```bash wget https://huggingface.co/FM-1976/gemma-2-2b-it-Q5_K_M-GGUF/resolve/main/gemma-2-2b-it-q5_k_m.gguf -OutFile gemma-2-2b-it-q5_k_m.gguf ``` ### Open your Python REPL #### Using chat_template ```python from llama_cpp import Llama nCTX = 8192 sTOPS = ['<eos>'] llm = Llama( model_path='gemma-2-2b-it-q5_k_m.gguf', temperature=0.24, n_ctx=nCTX, max_tokens=600, repeat_penalty=1.176, stop=sTOPS, verbose=False, ) messages = [ {"role": "user", "content": "Write me a poem about Machine Learning."}, ] response = llm.create_chat_completion( messages=messages, temperature=0.15, repeat_penalty= 1.178, stop=sTOPS, max_tokens=500) print(response['choices'][0]['message']['content']) ``` #### Using create_completion ```python from llama_cpp import Llama nCTX = 8192 sTOPS = ['<eos>'] llm = Llama( model_path='gemma-2-2b-it-q5_k_m.gguf', temperature=0.24, n_ctx=nCTX, max_tokens=600, repeat_penalty=1.176, stop=sTOPS, verbose=False, ) prompt = 'Explain Science in one sentence.' template = f'''<bos><start_of_turn>user {prompt}<end_of_turn> <start_of_turn>model <end_of_turn>''' res = llm.create_completion(prompt,temperature=0.15, max_tokens=500,repeat_penalty=1.178, stop=['<eos>']) print(res['choices'][0]['text']) ``` ### Streaming text llama-cpp-python allows you to also stream text during the inference<br> Tokens are decoded and printed soon after gneration is done. You don't have to wait until the entire inference is done. <br><br> You can use both `create_chat_completion()` and `create_completion()` methods. <br> #### Streaming with `create_chat_completion()` method ```python import datetime from llama_cpp import Llama nCTX = 8192 sTOPS = ['<eos>'] llm = Llama( model_path='gemma-2-2b-it-q5_k_m.gguf', temperature=0.24, n_ctx=nCTX, max_tokens=600, repeat_penalty=1.176, stop=sTOPS, verbose=False, ) fisrtround=0 full_response = '' message = [{'role':'user','content':'what is science?'}] start = datetime.datetime.now() for chunk in llm.create_chat_completion( messages=message, temperature=0.15, repeat_penalty= 1.31, stop=['<eos>'], max_tokens=500, stream=True,): try: if chunk["choices"][0]["delta"]["content"]: if fisrtround==0: print(chunk["choices"][0]["delta"]["content"], end="", flush=True) full_response += chunk["choices"][0]["delta"]["content"] ttftoken = datetime.datetime.now() - start fisrtround = 1 else: print(chunk["choices"][0]["delta"]["content"], end="", flush=True) full_response += chunk["choices"][0]["delta"]["content"] except: pass first_token_time = ttftoken.total_seconds() print(f'Time to first token: {first_token_time:.2f} seconds') ``` #### Streaming with `create_completion()` method ```python import datetime from llama_cpp import Llama nCTX = 8192 sTOPS = ['<eos>'] llm = Llama( model_path='gemma-2-2b-it-q5_k_m.gguf', temperature=0.24, n_ctx=nCTX, max_tokens=600, repeat_penalty=1.176, stop=sTOPS, verbose=False, ) fisrtround=0 full_response = '' prompt = 'Explain Science in one sentence.' template = f'''<bos><start_of_turn>user {prompt}<end_of_turn> <start_of_turn>model <end_of_turn>''' start = datetime.datetime.now() for chunk in llm.create_completion( prompt, temperature=0.15, repeat_penalty= 1.78, stop=['<eos>'], max_tokens=500, stream=True,): if fisrtround==0: print(chunk["choices"][0]["text"], end="", flush=True) full_response += chunk["choices"][0]["text"] ttftoken = datetime.datetime.now() - start fisrtround = 1 else: print(chunk["choices"][0]["text"], end="", flush=True) full_response += chunk["choices"][0]["text"] first_token_time = ttftoken.total_seconds() print(f'Time to first token: {first_token_time:.2f} seconds') ``` ### Further exploration You can also serve the model with an OpenAI compliant API server<br> This can be done both with `llama-cpp-python[server]` and `llamafile`.
mav23/Llama3-OpenBioLLM-8B-GGUF
mav23
2024-10-13T01:46:59Z
49
0
null
[ "gguf", "llama-3", "llama", "Mixtral", "instruct", "finetune", "chatml", "DPO", "RLHF", "gpt4", "distillation", "en", "arxiv:2305.18290", "arxiv:2303.13375", "arxiv:2212.13138", "arxiv:2305.09617", "arxiv:2402.07023", "base_model:meta-llama/Meta-Llama-3-8B", "base_model:quantized:meta-llama/Meta-Llama-3-8B", "license:llama3", "endpoints_compatible", "region:us" ]
null
2024-10-13T00:55:51Z
--- base_model: meta-llama/Meta-Llama-3-8B tags: - llama-3 - llama - Mixtral - instruct - finetune - chatml - DPO - RLHF - gpt4 - distillation model-index: - name: OpenBioLLM-8B results: [] license: llama3 language: - en widget: - example_title: OpenBioLLM-8B messages: - role: system content: >- You are an expert and experienced from the healthcare and biomedical domain with extensive medical knowledge and practical experience. - role: user content: How long does it take for newborn jaundice to go away? output: text: >- Newborn jaundice, also known as neonatal jaundice, is a common condition in newborns where the yellowing of the skin and eyes occurs due to an elevated level of bilirubin in the blood. Bilirubin is a yellow pigment that forms when red blood cells break down. In most cases, newborn jaundice resolves on its own without any specific treatment. The duration of newborn jaundice can vary depending on several factors such as the underlying cause, gestational age at birth, and individual variations in bilirubin metabolism. Here are some general guidelines: 1. Physiological jaundice: This is the most common type of newborn jaundice and usually appears within 24-72 hours after birth. It tends to peak between the second and fifth day of life and gradually improves over the next week or two. By the time the baby is one week old, the jaundice should have mostly resolved. 2. Breast milk jaundice: This type of jaundice occurs in breastfed babies and may appear later than physiological jaundice, typically between the fifth and fourteenth day of life. It tends to persist for a longer duration but usually resolves within six weeks after birth. 3. Pathological jaundice: This type of jaundice is less common and occurs due to an underlying medical condition that affects bilirubin metabolism or liver function. The duration of pathological jaundice depends on the specific cause and may require treatment. It's important for parents to monitor their newborn's jaundice closely and seek medical advice if the jaundice progresses rapidly, becomes severe, or is accompanied by other symptoms such as poor feeding, lethargy, or excessive sleepiness. In these cases, further evaluation and management may be necessary. Remember that each baby is unique, and the timing of jaundice resolution can vary. If you have concerns about your newborn's jaundice, it's always best to consult with a healthcare professional for personalized advice and guidance. --- <div align="center"> <img width="260px" src="https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/BrQCb95lmEIFz79QAmoNA.png"></div> ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/2FhDh8NDvMl7iSxbQz9BP.png) <div align="center"> <h1>Advancing Open-source Large Language Models in Medical Domain</h1> </div> <p align="center" style="margin-top: 0px;"> <a href="https://colab.research.google.com/drive/1F5oV20InEYeAJGmBwYF9NM_QhLmjBkKJ?usp=sharing"> <img src="https://colab.research.google.com/assets/colab-badge.svg" alt="OpenChat Logo" style="width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 10px; margin-top: 0px; margin-bottom: 0px;"/> <span class="link-text" style=" margin-right: 5px;">Online Demo</span> </a> | <a href="https://github.com/openlifescience-ai"> <img src="https://github.githubassets.com/assets/GitHub-Mark-ea2971cee799.png" alt="GitHub Logo" style="width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 5px; margin-top: 0px; margin-bottom: 0px;"/> <span class="link-text" style=" margin-right: 5px;">GitHub</span> </a> | <a href="#"> <img src="https://github.com/alpayariyak/openchat/blob/master/assets/arxiv-logomark-small-square-border.png?raw=true" alt="ArXiv Logo" style="width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 5px; margin-top: 0px; margin-bottom: 0px;"/> <span class="link-text" style="margin-right: 5px;">Paper</span> </a> | <a href="https://discord.gg/A5Fjf5zC69"> <img src="https://cloud.githubusercontent.com/assets/6291467/26705903/96c2d66e-477c-11e7-9f4e-f3c0efe96c9a.png" alt="Discord Logo" style="width:20px; vertical-align: middle; display: inline-block; margin-right: 5px; margin-left: 5px; margin-top: 0px; margin-bottom: 0px;"/> <span class="link-text">Discord</span> </a> </p> ![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/KGmRE5w2sepNtwsEu8t7K.jpeg) Introducing OpenBioLLM-8B: A State-of-the-Art Open Source Biomedical Large Language Model OpenBioLLM-8B is an advanced open source language model designed specifically for the biomedical domain. Developed by Saama AI Labs, this model leverages cutting-edge techniques to achieve state-of-the-art performance on a wide range of biomedical tasks. 🏥 **Biomedical Specialization**: OpenBioLLM-8B is tailored for the unique language and knowledge requirements of the medical and life sciences fields. It was fine-tuned on a vast corpus of high-quality biomedical data, enabling it to understand and generate text with domain-specific accuracy and fluency. 🎓 **Superior Performance**: With 8 billion parameters, OpenBioLLM-8B outperforms other open source biomedical language models of similar scale. It has also demonstrated better results compared to larger proprietary & open-source models like GPT-3.5 and Meditron-70B on biomedical benchmarks. 🧠 **Advanced Training Techniques**: OpenBioLLM-8B builds upon the powerful foundations of the **Meta-Llama-3-8B** and [Meta-Llama-3-8B](meta-llama/Meta-Llama-3-8B) models. It incorporates the DPO dataset and fine-tuning recipe along with a custom diverse medical instruction dataset. Key components of the training pipeline include: <div align="center"> <img width="1200px" src="https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/oPchsJsEpQoGcGXVbh7YS.png"> </div> - **Policy Optimization**: [Direct Preference Optimization: Your Language Model is Secretly a Reward Model (DPO)](https://arxiv.org/abs/2305.18290) - **Ranking Dataset**: [berkeley-nest/Nectar](https://huggingface.co/datasets/berkeley-nest/Nectar) - **Fine-tuning dataset**: Custom Medical Instruct dataset (We plan to release a sample training dataset in our upcoming paper; please stay updated) This combination of cutting-edge techniques enables OpenBioLLM-8B to align with key capabilities and preferences for biomedical applications. ⚙️ **Release Details**: - **Model Size**: 8 billion parameters - **Quantization**: Optimized quantized versions available [Here](https://huggingface.co/aaditya/OpenBioLLM-Llama3-8B-GGUF) - **Language(s) (NLP):** en - **Developed By**: [Ankit Pal (Aaditya Ura)](https://aadityaura.github.io/) from Saama AI Labs - **License:** Meta-Llama License - **Fine-tuned from models:** [meta-llama/Meta-Llama-3-8B](meta-llama/Meta-Llama-3-8B) - **Resources for more information:** - Paper: Coming soon The model can be fine-tuned for more specialized tasks and datasets as needed. OpenBioLLM-8B represents an important step forward in democratizing advanced language AI for the biomedical community. By leveraging state-of-the-art architectures and training techniques from leading open source efforts like Llama-3, we have created a powerful tool to accelerate innovation and discovery in healthcare and the life sciences. We are excited to share OpenBioLLM-8B with researchers and developers around the world. ### Use with transformers **Important: Please use the exact chat template provided by Llama-3 instruct version. Otherwise there will be a degradation in the performance. The model output can be verbose in rare cases. Please consider setting temperature = 0 to make this happen less.** See the snippet below for usage with Transformers: ```python import transformers import torch model_id = "aaditya/OpenBioLLM-Llama3-8B" pipeline = transformers.pipeline( "text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device="auto", ) messages = [ {"role": "system", "content": "You are an expert and experienced from the healthcare and biomedical domain with extensive medical knowledge and practical experience. Your name is OpenBioLLM, and you were developed by Saama AI Labs. who's willing to help answer the user's query with explanation. In your explanation, leverage your deep medical expertise such as relevant anatomical structures, physiological processes, diagnostic criteria, treatment guidelines, or other pertinent medical concepts. Use precise medical terminology while still aiming to make the explanation clear and accessible to a general audience."}, {"role": "user", "content": "How can i split a 3mg or 4mg waefin pill so i can get a 2.5mg pill?"}, ] prompt = pipeline.tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True ) terminators = [ pipeline.tokenizer.eos_token_id, pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>") ] outputs = pipeline( prompt, max_new_tokens=256, eos_token_id=terminators, do_sample=True, temperature=0.0, top_p=0.9, ) print(outputs[0]["generated_text"][len(prompt):]) ``` ## **Training procedure** ### **Training hyperparameters** <details> <summary>Click to see details</summary> - learning_rate: 0.0002 - lr_scheduler: cosine - train_batch_size: 12 - eval_batch_size: 8 - GPU: H100 80GB SXM5 - num_devices: 1 - optimizer: adamw_bnb_8bit - lr_scheduler_warmup_steps: 100 - num_epochs: 4 </details> ### **Peft hyperparameters** <details> <summary>Click to see details</summary> - adapter: qlora - lora_r: 128 - lora_alpha: 256 - lora_dropout: 0.05 - lora_target_linear: true -lora_target_modules: - q_proj - v_proj - k_proj - o_proj - gate_proj - down_proj - up_proj </details> ### **Training results** ### **Framework versions** - Transformers 4.39.3 - Pytorch 2.1.2+cu121 - Datasets 2.18.0 - Tokenizers 0.15.1 - Axolotl - Lm harness for evaluation # Benchmark Results 🔥 OpenBioLLM-8B demonstrates superior performance compared to larger models, such as GPT-3.5, Meditron-70B across 9 diverse biomedical datasets, achieving state-of-the-art results with an average score of 72.50%, despite having a significantly smaller parameter count. The model's strong performance in domain-specific tasks, such as Clinical KG, Medical Genetics, and PubMedQA, highlights its ability to effectively capture and apply biomedical knowledge. 🚨 The GPT-4, Med-PaLM-1, and Med-PaLM-2 results are taken from their official papers. Since Med-PaLM doesn't provide zero-shot accuracy, we are using 5-shot accuracy from their paper for comparison. All results presented are in the zero-shot setting, except for Med-PaLM-2 and Med-PaLM-1, which use 5-shot accuracy. | | Clinical KG | Medical Genetics | Anatomy | Pro Medicine | College Biology | College Medicine | MedQA 4 opts | PubMedQA | MedMCQA | Avg | |--------------------|-------------|------------------|---------|--------------|-----------------|------------------|--------------|----------|---------|-------| | **OpenBioLLM-70B** | **92.93** | **93.197** | **83.904** | 93.75 | 93.827 | **85.749** | 78.162 | 78.97 | **74.014** | **86.05588** | | Med-PaLM-2 (5-shot) | 88.3 | 90 | 77.8 | **95.2** | 94.4 | 80.9 | **79.7** | **79.2** | 71.3 | 84.08 | | **GPT-4** | 86.04 | 91 | 80 | 93.01 | **95.14** | 76.88 | 78.87 | 75.2 | 69.52 | 82.85 | | Med-PaLM-1 (Flan-PaLM, 5-shot) | 80.4 | 75 | 63.7 | 83.8 | 88.9 | 76.3 | 67.6 | 79 | 57.6 | 74.7 | | **OpenBioLLM-8B** | 76.101 | 86.1 | 69.829 | 78.21 | 84.213 | 68.042 | 58.993 | 74.12 | 56.913 | 72.502 | | Gemini-1.0 | 76.7 | 75.8 | 66.7 | 77.7 | 88 | 69.2 | 58 | 70.7 | 54.3 | 70.79 | | GPT-3.5 Turbo 1106 | 74.71 | 74 | 72.79 | 72.79 | 72.91 | 64.73 | 57.71 | 72.66 | 53.79 | 66 | | Meditron-70B | 66.79 | 69 | 53.33 | 71.69 | 76.38 | 63 | 57.1 | 76.6 | 46.85 | 64.52 | | gemma-7b | 69.81 | 70 | 59.26 | 66.18 | 79.86 | 60.12 | 47.21 | 76.2 | 48.96 | 64.18 | | Mistral-7B-v0.1 | 68.68 | 71 | 55.56 | 68.38 | 68.06 | 59.54 | 50.82 | 75.4 | 48.2 | 62.85 | | Apollo-7B | 62.26 | 72 | 61.48 | 69.12 | 70.83 | 55.49 | 55.22 | 39.8 | 53.77 | 60 | | MedAlpaca-7b | 57.36 | 69 | 57.04 | 67.28 | 65.28 | 54.34 | 41.71 | 72.8 | 37.51 | 58.03 | | BioMistral-7B | 59.9 | 64 | 56.5 | 60.4 | 59 | 54.7 | 50.6 | 77.5 | 48.1 | 57.3 | | AlpaCare-llama2-7b | 49.81 | 49 | 45.92 | 33.82 | 50 | 43.35 | 29.77 | 72.2 | 34.42 | 45.36 | | ClinicalGPT | 30.56 | 27 | 30.37 | 19.48 | 25 | 24.27 | 26.08 | 63.8 | 28.18 | 30.52 | <div align="center"> <img width="1600px" src="https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/_SzdcJSBjZyo8RS1bTEkP.png"> </div> ## Detailed Medical Subjectwise accuracy ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/UXF-V0col0Z0sS6BGPBkE.png) # Use Cases & Examples 🚨 **Below results are from the quantized version of OpenBioLLM-70B** # Summarize Clinical Notes OpenBioLLM-70B can efficiently analyze and summarize complex clinical notes, EHR data, and discharge summaries, extracting key information and generating concise, structured summaries ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/xdwdBgOxNi_TfML0hKlI8.png) # Answer Medical Questions OpenBioLLM-70B can provide answers to a wide range of medical questions. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/zO95GlwOQEZqCKQF69mE6.png) ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/OKBczKw7gWeW5xsuDpc27.png) <details> <summary>Click to see details</summary> ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/eJGHT5khppYvJb8fQ-YW4.png) ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/Cnbwrqa_-ORHRuNRC2P6Y.png) ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/J9DhdcvukAc9mnnW9fj2C.png) </details> # Clinical Entity Recognition OpenBioLLM-70B can perform advanced clinical entity recognition by identifying and extracting key medical concepts, such as diseases, symptoms, medications, procedures, and anatomical structures, from unstructured clinical text. By leveraging its deep understanding of medical terminology and context, the model can accurately annotate and categorize clinical entities, enabling more efficient information retrieval, data analysis, and knowledge discovery from electronic health records, research articles, and other biomedical text sources. This capability can support various downstream applications, such as clinical decision support, pharmacovigilance, and medical research. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/_69BW4k9LVABFwtxixL45.png) ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/DKy5wYCoPhoPPUc1-x8_J.png) ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/7WD9zCCBZT4-4XlfnIQjl.png) # Biomarkers Extraction ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/ZttoM4AiteT7gFYVhjIpN.png) # Classification OpenBioLLM-70B can perform various biomedical classification tasks, such as disease prediction, sentiment analysis, medical document categorization ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/Bf5MW1d75qT-1F_TR_hC0.png) # De-Identification OpenBioLLM-70B can detect and remove personally identifiable information (PII) from medical records, ensuring patient privacy and compliance with data protection regulations like HIPAA. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/5f3fe13d79c1ba4c353d0c19/hKX4kzm--Tw5bj6K78msy.png) **Advisory Notice!**  While OpenBioLLM-70B & 8B leverages high-quality data sources, its outputs may still contain inaccuracies, biases, or misalignments that could pose risks if relied upon for medical decision-making without further testing and refinement. The model's performance has not yet been rigorously evaluated in randomized controlled trials or real-world healthcare environments. Therefore, we strongly advise against using OpenBioLLM-70B & 8B for any direct patient care, clinical decision support, or other professional medical purposes at this time. Its use should be limited to research, development, and exploratory applications by qualified individuals who understand its limitations. OpenBioLLM-70B & 8B are intended solely as a research tool to assist healthcare professionals and should never be considered a replacement for the professional judgment and expertise of a qualified medical doctor. Appropriately adapting and validating OpenBioLLM-70B & 8B for specific medical use cases would require significant additional work, potentially including: - Thorough testing and evaluation in relevant clinical scenarios - Alignment with evidence-based guidelines and best practices - Mitigation of potential biases and failure modes - Integration with human oversight and interpretation - Compliance with regulatory and ethical standards Always consult a qualified healthcare provider for personal medical needs. # Citation If you find OpenBioLLM-70B & 8B useful in your work, please cite the model as follows: ``` @misc{OpenBioLLMs, author = {Ankit Pal, Malaikannan Sankarasubbu}, title = {OpenBioLLMs: Advancing Open-Source Large Language Models for Healthcare and Life Sciences}, year = {2024}, publisher = {Hugging Face}, journal = {Hugging Face repository}, howpublished = {\url{https://huggingface.co/aaditya/OpenBioLLM-Llama3-70B}} } ``` The accompanying paper is currently in progress and will be released soon. <div align="center"> <h2> 💌 Contact </h2> </div> We look forward to hearing you and collaborating on this exciting project! **Contributors:** - [Ankit Pal (Aaditya Ura)](https://aadityaura.github.io/) [aadityaura at gmail dot com] - Saama AI Labs - Note: I am looking for a funded PhD opportunity, especially if it fits my Responsible Generative AI, Multimodal LLMs, Geometric Deep Learning, and Healthcare AI skillset. # References We thank the [Meta Team](meta-llama/Meta-Llama-3-70B-Instruct) for their amazing models! Result sources - [1] GPT-4 [Capabilities of GPT-4 on Medical Challenge Problems] (https://arxiv.org/abs/2303.13375) - [2] Med-PaLM-1 [Large Language Models Encode Clinical Knowledge](https://arxiv.org/abs/2212.13138) - [3] Med-PaLM-2 [Towards Expert-Level Medical Question Answering with Large Language Models](https://arxiv.org/abs/2305.09617) - [4] Gemini-1.0 [Gemini Goes to Med School](https://arxiv.org/abs/2402.07023)
josedossantos/tree-txtInteiroTeorLimpo-bertimbau
josedossantos
2024-10-13T01:37:16Z
5
0
sentence-transformers
[ "sentence-transformers", "safetensors", "bert", "sentence-similarity", "feature-extraction", "generated_from_trainer", "dataset_size:4327", "loss:ContrastiveLoss", "arxiv:1908.10084", "base_model:neuralmind/bert-large-portuguese-cased", "base_model:finetune:neuralmind/bert-large-portuguese-cased", "autotrain_compatible", "text-embeddings-inference", "endpoints_compatible", "region:us" ]
sentence-similarity
2024-10-13T01:36:05Z
--- base_model: neuralmind/bert-large-portuguese-cased datasets: [] language: [] library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:4327 - loss:ContrastiveLoss widget: - source_sentence: PROJETO DE LEI No DE 2015 Do Sr. Veneziano Vital do Rêgo Dispõe sobre os procedimentos aplicáveis em benefício do consumidor de produtos eletrônicos ou eletrodomésticos, em caso de obsolescência do produto antes do término de sua vida útil. O Congresso Nacional decreta Art. 1º A oferta e apresentação de produtos eletrônicos e eletrodomésticos deve fornecer, além das informações obrigatórias estabelecidas na legislação aplicável, a vida útil estimada do produto introduzido no mercado de consumo. Art. 2º Em caso de superveniente obsolescência, sem culpa do consumidor, do produto eletrônico ou eletrodoméstico antes do término de seu prazo de vida útil, pode o consumidor exigir, alternativamente e à sua escolha, que o fornecedor, no prazo de 7 sete dias úteis I restitua a quantia paga, monetariamente atualizada II substitua o produto por outro da mesma espécie ou por similar de melhor qualidade Art. 3º O não cumprimento do prazo estabelecido no artigo anterior sujeita o fornecedor, sem prejuízo das penalidades cabíveis por força da legislação de proteção e defesa do consumidor, à multa correspondente a 30 trinta por cento do valor monetariamente atualizado de aquisição do produto, que deverá ser revertida, imediatamente, ao consumidor. 2 Art. 4º Esta Lei entra em vigor na data de sua publicação oficial. JUSTIFICAÇÃO O presente projeto de lei objetiva enfrentar a cada vez mais usual prática de fabricantes de produtos eletrônicos e eletrodomésticos de, por variados artifícios e em curto espaço de tempo, tornar obsoletos os bens adquiridos pelo consumidor. Nos últimos anos, todos temos acompanhado as agressivas campanhas de marketing que incutem nos consumidores uma lógica de status e de consumo instantâneo. Essas técnicas objetivam conduzir os consumidores à compreensão de que aquele produto recém adquirido e cujas qualidades e possibilidades há pouco foram tão fortemente explicitadas já se encontra ultrapassado e que, agora, somente um novo e mais desenvolvido produto poderia atender as necessidades do consumidor. A par dessa obsolescência de desejabilidade, fruto de publicidade e propagandas exageradas, prolifera também em nossa sociedade de consumo a obsolescência de qualidade e de função, causada pelos próprios fabricantes. Esses, de modo deliberado e planejado, colocam no mercado produtos de reduzida durabilidade e utilidade, compelindo o consumidor, em tempo muito mais curto do que o esperado para aquele tipo de produto, a substituí lo. Entendemos que nosso projeto ao obrigar a informação da vida útil dos produtos e ao exigir a substituição do produto ou a restituição do valor pago em caso de obsolescência contribuirá para reprimir essa conduta tão fortemente prejudicial aos consumidores. 3 Contamos com a colaboração dos ilustres Pares para o aperfeiçoamento e aprovação da presente proposição, considerando que trará inegáveis benefícios ao consumidor brasileiro. Sala das Sessões, em de de 2015. Deputado VENEZIANO VITAL DO RÊGO 2015_18501 sentences: - PROJETO DE LEI Nº DE 2011 Do Sr. ARTHUR LIRA Altera o art. 18 da Lei nº 8. 078, de 1990, para facultar ao consumidor o direito de exigir a substituição produto viciado, no prazo de sete dias, contado a partir da data da entrega. O Congresso Nacional decreta Art. 1º Esta lei faculta ao consumidor o direito de exigir a substituição de produto viciado, no prazo de até sete dias, contado a partir da data da entrega. Art. 2º O caput do art. 18 da Lei nº 8. 078, de 11 de setembro de 1990, passa a vigorar com a seguinte redação Art. 18. Os fornecedores de produtos de consumo duráveis ou não duráveis respondem solidariamente pelos vícios de qualidade ou quantidade que os tornem impróprios ou inadequados ao consumo a que se destinam ou lhes diminuam o valor, assim como por aqueles decorrentes da disparidade com as indicações constantes do recipiente, da embalagem, rotulagem ou mensagem publicitária, respeitadas as variações decorrentes de sua natureza, podendo o consumidor exigir a substituição do produto viciado, no prazo de até sete dias, contado a partir da data da entrega e, após esse prazo, exigir a substituição das partes viciadas. NR Art. 3º Esta lei entra em vigor na data de sua publicação. 2 JUSTIFICAÇÃO O atual estágio de desenvolvimento científico e tecnológico em que nos encontramos e a consciência da maioria das empresas de que devem primar pela qualidade de seus produtos são fenômenos compatíveis com o elevado grau de exigência do consumidor hodierno. Na época que corre, marcada pelo intenso uso de controles e programas de qualidade nas indústrias, não é mais possível aceitar a distribuição de produtos defeituosos no mercado de consumo. Embora seja estatisticamente natural que alguma unidade de produto escape ao controle de qualidade e chegue ao mercado com defeito, não é possível admitir que o consumidor arque com prejuízo causado, unicamente, por falha do fornecedor. Vale ressaltar a existência de legislação protetora contra produto defeituoso. O artigo 26 da Lei nº 8. 078, de 1990, concede prazo de trinta dias para o consumidor exercer seu direito de reclamar por vícios aparentes ou de fácil constatação em produtos não duráveis e o prazo de noventa dias, quando se tratar de produtos duráveis. Esse direito é conhecido como prazo de garantia legal e independe de qualquer contrato de garantia oferecido pelo fornecedor. Portanto, no prazo de trinta ou noventa dias, a depender do tipo de produto, o consumidor pode acionar a proteção prevista no artigo 18 do Código de Defesa do Consumidor CDC, que, por sua vez, garante o direito à substituição das partes viciadas de qualquer produto, no prazo de até trinta dias, assim como faculta o direito à substituição do produto se, em trinta dias, o problema não for solucionado. Além disso, o 3º do artigo 18 do citado Código estabelece que, excepcionalmente, quando se tratar de produto essencial, o consumidor tem direito à substituição imediata do produto com defeito. Entretanto, apesar da existência dessa legislação protetora, constatamos que ela não atende satisfatoriamente a defesa do consumidor, por vários motivos. Por exemplo, não encontramos no CDC definição clara do que seja produto essencial. Um telefone celular seria um produto essencial Uma geladeira seria um produto essencial Parece que não, porque as notícias que nos chegam é que sempre que o consumidor reclama de vícios nesse tipo de produto, ainda que dentro do prazo de garantia 3 legal, não acontece sua substituição por outro, mas encaminha se o produto viciado à assistência técnica, privando o consumidor de utilizá lo. Portanto, para nós é claro que os atuais dispositivos da Lei nº 8. 078, de 1990, são incapazes de oferecer uma proteção adequada ao consumidor contra os produtos viciados que são distribuídos no mercado de consumo. A iniciativa que ora oferecemos à apreciação dos ilustres Pares busca estender a todos os produtos o benefício da substituição imediata, hoje restrito aos produtos considerados essenciais. Porém, fixa um prazo mais curto, de sete dias, para o consumidor reivindicar a substituição dos produtos em geral, enquanto mantém o prazo em vigor, de trinta dias, para os considerados essenciais. Pelas razões acima expostas, solicitamos o decisivo apoio dos nobres Pares para a aprovação desta iniciativa. Sala das Sessões, em de de 2011. Deputado ARTHUR LIRA 2011_12670 - PROJETO DE LEI No DE 2007 Do Sr. OTAVIO LEITE Dispõe sobre a utilização das áreas públicas de unidades de conservação ambiental integrantes do S. N. U. C. Sistema Nacional de Unidades de Conservação de unidades militares e prédios públicos em geral, particularmente os destinados às unidades educacionais, por grupos oficiais de Escoteiros e Bandeirantes e dá outras providências. O Congresso Nacional decreta Art. 1º As áreas públicas de unidades de conservação ambiental integrantes do S. N. U. C. Sistema Nacional de Unidades de Conservação de unidades militares e de prédios públicos em geral, particularmente os destinados às unidades educacionais, poderão ser disponibilizadas, em horários e espaços compatíveis com seus respectivos funcionamentos regulares, para a realização de atividades desenvolvidas por grupos oficiais de Escoteiros e Bandeirantes. Parágrafo Único Não haverá de vínculo entre matrícula nas unidades educacionais e a adesão a determinado grupo de Bandeirantes ou Escoteiros. Art. 2º O Poder Público, sempre que possível, garantirá a infra estrutura adequada dos locais referidos no artigo anterior, com equipamentos sanitários e sistemas de energia, iluminação e segurança quando do desenvolvimento das atividades pelos grupos de Escoteiros e Bandeirantes, respectivamente filiados à União dos Escoteiros do Brasil e ou à Federação de Bandeirantes do Brasil. Art. 3º Para a consecução dos objetivos desta Lei, os grupos oficiais de Escoteiros e Bandeirantes deverão requerer o espaço a ser utilizado, diretamente aos titulares do órgão unidade no qual pretendam implantar suas atividades, detalhando horários e seus programas de trabalho, para fins de avaliação e autorização pertinente. 2 Art. 4º A autorização de que trata o artigo anterior será concedida a título precário, ficando os grupos oficiais de Escoteiros e Bandeirantes responsáveis pela conservação e manutenção dos espaços cedidos para suas atividades. Parágrafo Único. A inobservância do caput deste artigo implicará em suspensão da disponibilização dos espaços aludidos. Art. 5º Esta lei entra em vigor na data da sua publicação. JUSTIFICAÇÃO Iniciaram se os Movimentos Escoteiro e Bandeirante no início do século passado por iniciativa, de Robert Baden Powell e de sua irmã Agnes, na Inglaterra. Da idéia inicial, consubstanciada no livro Escotismo para rapazes editado em 1908, que vê, no scout, aquele que vai à frente perscrutando o terreno, formaram se movimentos infanto juvenis de educação não formal mundiais, propagadores do altruísmo, da lealdade, da fraternidade, da responsabilidade, do respeito a si mesmo e ao próximo e da disciplina perseverante, baseados em um sistema de valores que visam a desenvolver o senso crítico, a criatividade, a participação, o contato com a natureza e a espiritualidade. A metodologia dos Movimentos Bandeirante e Escoteiro está calcada em cinco vertentes, que são a vida em equipe, o compromisso com o desenvolvimento pessoal e social, a progressão pessoal e do grupo, o contato respeitoso e respeitador com a natureza e a participação ativa na comunidade, que se espelham no compromisso que marca o momento de adesão do participante ao grupo Farei o melhor possível para o meu aperfeiçoamento pessoal, participar de maneira ativa e responsável na comunidade, buscar a fé e vivenciar os princípios contidos no código princípios, esses, que se traduzem por fraternidade universal, dever para com os outros, autodisciplina e respeito à natureza, procurando superar o egoísmo, o pessimismo, as diferenças religiosas, os problemas de classe e as questões sociais O compromisso formulado pela própria pessoa também é reconhecido como válido pela Federação de Bandeirantes do Brasil, desde que reflita a verdade e o compromisso pessoais com o crescimento sadio e a disposição em participar do engajamento social da associação não só na 3 comunidade a que o grupo pertencer, como na casa planetária, haja vista os vários programas de capacitação e intercâmbio que as referidas associações propiciam, visando a aproximar crianças e jovens do mundo todo, desde as antigas Caixas de Correspondência que estimulavam a correspondência além fronteiras, ao intercâmbio hoje facilitado pela internet. Surgiram, no Brasil, desde o início do século passado, foram se desenvolvendo, espalhando se pelo País e deram origem à Federação de Bandeirantes do Brasil e à União dos Escoteiros do Brasil, que se filiam, respectivamente, à Associação Mundial de Bandeirantes e à Organização Mundial do Movimento Escoteiro, todos tendo a natureza jurídica de associações civis, portanto sem fins lucrativos, também reconhecidas, no Brasil, como associações de utilidade pública. A metodologia dos dois movimentos, a ênfase constante na responsabilidade de cada um pelas escolhas individuais e coletivas feitas e o compromisso decorrente de todos pelos resultados conquistados a equipe assumindo os resultados e as conseqüências talvez tenha sido seu diferencial ao longo dos anos, sua contribuição à formação grupos em que lealdade e solidariedade são valores vivenciados. No caso do Movimento Bandeirante, há, também, de se ressaltar a sua contribuição para a educação e emancipação femininas ao longo do século passado no mundo e no Brasil uma menina ou jovem que atuava em sua comunidade de forma participativa e ousava sair para acampar com seu grupo, montando a própria barraca e a infra estrutura do acampamento, pensava o universo feminino de uma forma diferenciada. Exemplo disso emblemático é o do voto feminino se, na equipe bandeirante, os projetos eram discutidos e votados e todos eram responsáveis pelos seus resultados, também, nos países, tinham as mulheres o direito de exercer o voto em suas comunidades. A criatividade e a participação ativa destacam se, no Brasil, em expoentes, que vão de Maria Clara Machado, no teatro, a sua filha Ana Maria, na literatura infantil, a Salete Maria Polita Maccalóz, nos novos tempos do judiciário brasileiro. São movimentos de educação não formal, não fazem distinção de credo, gênero ou etnia e não se vinculam a partidos políticos, embora ofereçam todas as oportunidades para o pleno desenvolvimento da 4 cidadania ativa, através das diferentes formas de capacitação pessoal e da equipe, tanto sob o ponto de vista físico, com atividades que melhorem o condicionamento e a agilidade física e mental, como intelectual, ético e social, que se refletem tanto no cuidado com a natureza como na participação ativa na vida comunitária através de trabalhos voluntários, sempre com o objetivo de ensinar a pescar, sem se limitar à doação do peixe, tendo tido atuação destacada em momentos emergenciais de ajuda humanitária. No que concerne à educação ambiental, hoje prevista em lei, convém lembrar que os dois movimentos têm desenvolvido projetos sistemáticos a respeito desde o início do século passado. No início da década de 60, o Projeto Natureza da FBB, visando, especificamente, à redução da poluição e do consumo de aerossóis para redução do efeito estufa, época em que meio ambiente não era ainda um modismo e falar em controle da poluição do ar soava a excrescência técnica e jurídica. É comum a participação de escoteiros em iniciativas de preservação de Parques Nacionais com orientação de princípios de conduta aos visitantes, auxílio ao manejo de trilhas e reflorestamento. No Rio de Janeiro os escoteiros mantêm e preservam um trecho de Mata Atlântica de 40 hectares, no Município de Guapimirim, outrora ligado à floresta que recobre o Maciço da Serra dos Órgãos. A reserva tem sido visitada freqüentemente ao longo dos últimos 30 anos, por alunos e pesquisadores de universidades públicas e particulares dando ensejo ao aperfeiçoamento de saberes ligados às ciências biológicas. São, todas essas, razões pelas quais, em nosso País, os chamados Movimentos Escoteiro e Bandeirante destacaram se como organizações não governamentais de educação não formal, sem fins lucrativos e de reconhecida utilidade pública. A metodologia dos dois movimentos é aplicada aos jovens através de brincadeiras, jogos e exercícios, de uma forma de vida saudável, repleta de bom humor e camaradagem leal e solidária, onde a aprendizagem é uma constante, que começa na formação do consenso coletivo, através de tomadas de decisões conjuntas na equipe e da respectiva avaliação franca dos resultados., em pleno exercício da democracia. Não são esses instrumentos úteis e aplicáveis a qualquer esfera e tempo de vida, seja na solução de problemas, seja na improvisação conseqüente, no conviver leal 5 e solidário, na criação de mecanismos de ajuda não dependizante, na confraternização despojada As atividades dos grupos bandeirantes e escoteiros são supervisionados por adultos voluntários, que se tenham submetido ao processo de capacitação pertinente de uma ou outra associação. Os dois movimentos são reconhecidos pela Organização das Nações Unidas como uma das maiores organizações internacionais de educação, congregando mais de vinte milhões de crianças, adolescentes e jovens em 216 países de todos os continentes, bem como de adultos engajados que mantêm acesa sua capacidade de sonhar e sua disposição de agir para criar um mundo melhor. No Brasil de hoje, todavia, a realidade é que muitos grupos de escoteiros e bandeirantes deixam de existir por falta de apoio e incentivo e por não disporem de espaço físico para suas reuniões semanais, para a guarda de seus objetos, desenvolvimento de suas atividades e concretização de seus projetos. Diante da importância educacional e social dessas organizações, nos termos proclamados por Baden Powell, nos idos de 1908, quando afirmou, com perspicácia de educador absolutamente inovadora, na época Se quisermos que nossos jovens sejam felizes na vida, devemos fazer com que eles assimilem o costume de praticar o bem ao próximo, além de ensinar lhes a apreciar as coisas da natureza, urge incentivá los, como potencial instrumento de efetiva inclusão social que se tem revelado ao longo desse século conflagrado, inclusive como instrumento de efetiva construção de educação para a paz. Essa iniciativa legislativa visa, assim, a preencher uma lacuna material, propondo a utilização dos prédios públicos, especialmente das escolas, para serem compartilhadas nos horários que não estão atendendo aos alunos, em classes ou em atividades extra classe. Normalmente, as reuniões bandeirantes e escoteiras são nos finais de semana, quando as escolas estão fechadas e não são utilizadas para atividades curriculares ou extra curriculares, possibilitando, assim, um aproveitamento adicional dos espaços públicos e uma motivação a mais para que as crianças e jovens busquem um lazer educativo e sadio, em movimentos de educação e formação. 6 As escolas são, por natureza, agregadoras e localizadas em pontos de fácil acesso não só para a comunidade local, como próximas a pontos de parada de ônibus, com uma infra estrutura significativa quanto à serviços de telecomunicações, equipamentos sanitários, área de recreação, variedade de ambientes para atividades de grupo, tendo, inclusive, locais adequados ao içamento de bandeiras, uma dentre outras formas de educação cívica que, no aprendizado escoteiro e bandeirante, tem caráter de participação e flexibilidade, respeito à própria pátria e a dos outros. Esperamos, pois, contar com o apoio dos nobres Pares para esta iniciativa legislativa, que visa a promover a educação não formal, em uma de suas matizes mais relevantes. Sala das Sessões, em de maio de 2007. OTAVIO LEITE Deputado Federal - CÂMARA DOS DEPUTADOS Deputado Laudívio Carvalho PROJETO DE LEI Nº DE 2015 Do Sr. LAUDIVIO CARVALHO Acrescenta o artigo 19 A à Lei nº 10. 098, de 19 de dezembro de 2000, para obrigar as empresas ligadas à televisão e ao cinema, assim como outros meios de comunicação visual, a colocar legendas em todos os filmes, novelas e similares exibidos, principalmente aqueles que forem dublados. O Congresso Nacional decreta Art. 1º Esta Lei acrescenta o artigo 19 A à Lei nº 10. 098, de 19 de dezembro de 2000, que Estabelece normas gerais e critérios básicos para a promoção da acessibilidade das pessoas portadoras de deficiência ou com mobilidade reduzida, e dá outras providências para obrigar as empresas ligadas à televisão e ao cinema, assim como outros meios de comunicação visual, a colocar legendas em todos os filmes, novelas e similares exibidos, principalmente aqueles que forem dublados. Art. 2º Acrescente se o seguinte artigo 19 A à Lei nº 10. 098, de 19 de dezembro de 2000 Art. 19 A É obrigatório o uso de legendas ocultas ou abertas em toda exibição de filmes, novelas e similares, tanto nos cinemas quanto nas emissoras de televisão, bem como em outros meios de comunicação visual destinados ao público em geral, na forma da regulamentação. 1º A regulamentação definirá os prazos para a adoção da obrigatoriedade estabelecida no caput deste artigo, de forma CÂMARA DOS DEPUTADOS Deputado Laudívio Carvalho a que toda exibição contenha as legendas no prazo máximo de 2 dois anos da data da publicação desta lei. 2º Na regulamentação serão priorizados os filmes, novelas e similares que forem dublados. 3º A obrigatoriedade de que trata o caput também se aplica aos filmes, novelas e similares que tenham sido gravados antes da publicação desta Lei, mas que venham a ser exibidos em prazos posteriores aos definidos pela regulamentação em conformidade com o 1º deste artigo. NR Art. 3º O Poder Executivo regulamentará esta Lei no prazo de 90 noventa dias contados de sua publicação. Art. 4º Esta Lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO Nosso País apresenta um número significativo de pessoas com diversos estágios de deficiência auditiva. Além disso, nossa população de idosos, que em muitos casos também apresenta redução da capacidade de escutar bem, também tem sido crescente. O Poder Público, entretanto, tem negligenciado no cuidado desta importante parcela de nossa sociedade. Dessa forma, exclui, cada vez mais, estas pessoas do convívio social regular e de opções de lazer. É bastante frequente encontrarmos pessoas com deficiência auditiva recolhidas ao seu mundo sem som, fechadas em si pela falta de oportunidade, que, na verdade, é um direito de todos. As opções de diversão mais populares em nosso País são exatamente a televisão e o cinema. Ao excluirmos dessas opções nossa população idosa e nossos cidadãos com deficiência auditiva, pecamos duas vezes a primeira pela retirada de um direito fundamental, que é o acesso ao lazer e, a segunda, pela aceleração do processo de exclusão social e pelo agravamento do quadro de saúde total das pessoas. Basta de tanta desatenção A principal função do Poder Público é exatamente a garantia do bem estar de toda a população, sem CÂMARA DOS DEPUTADOS Deputado Laudívio Carvalho discriminação de espécie alguma. Ao apresentarmos este Projeto de Lei, buscamos reverter esta situação de isolamento. Não podemos mais esperar para que nossos idosos e as pessoas com deficiência auditiva tenham seus direitos reconhecidos. Certos de que o alcance desta medida irá, por um lado, minimizar a gritante situação de desconforto de milhões de pessoas, e, por outro, garantir a cidadania plena a todos os brasileiros, contamos com o necessário o apoio de todos os parlamentares para a rápida aprovação desta matéria. Sala das Sessões, em 26 de maio de 2015. Deputado LAUDIVIO CARVALHO PMDB MG 2015 8741. docx - source_sentence: PROJETO DE LEI Nº DE 2015 Do Sr. JHC Altera o Decreto Lei nº 5. 452 de 1 de maio de 1943, que Aprova a Consolidação das Leis do Trabalho, para acrescer o inciso X e parágrafo único ao Art. 473, autorizando a que os empregados que realizem trabalhado voluntário possam deixar de comparecer ao serviço por 02 dois dias a cada 12 doze meses. O Congresso Nacional decreta Art. 1º O Art. 473 do Decreto Lei nº 5. 452 de 1 de maio de 1943 o seguinte inciso X e parágrafo único Art. 473 X 2 dois dias consecutivos ou não, a cada 12 doze meses, em caso trabalho voluntário prestado por no mínimo 40 horas na forma da Lei 9. 608 1998 Parágrafo único o acesso ao direito que se refere o inciso X deste artigo será disciplinado por Portaria do Ministério do Trabalho. Art. 2º Esta Lei entrará em vigor na data da sua publicação. JUSTIFICAÇÃO O Projeto de Lei em tela visa integrar ainda mais os Trabalhadores submetidos à CLT à sociedade, por meio de contribuição pessoal via trabalho voluntário. Em um país carente como o Brasil, toda guinada rumo objetivos cívicos, culturais, educacionais, científicos, recreativos ou de assistência social, deve ser valorizada. Nesta toada, nada mais natural que os trabalhadores, que já prestam inestimável serviço ao país contribuindo com sua força de trabalho, atuem de forma mais umbilical junto à sociedade civil organizada, seja em trabalhos eventuais ou auxílio diante de alguma tragédia. Além disso, a iniciativa em moldura permitirá que os trabalhadores possam organizar campanhas de cunho voluntário. Sala das Sessões, de de 2015. Deputado JHC sentences: - PROJETO DE LEI Nº DE 2012 Do Sr. Vitor Penido Denomina Rodovia Deputado Hugo Aguiar a rodovia BR 352, no Estado de Minas Gerais. O Congresso Nacional decreta Art. 1º Fica denominada Rodovia Deputado Hugo Aguiar a rodovia BR 352 no Estado de Minas Gerais, desde a divisa com o Estado de Goiás até o Município de Pará de Minas. Art. 2º Esta Lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO Com este projeto de lei, pretendemos homenagear o saudoso Deputado Hugo Aguiar, pela atribuição de seu nome ao traçado mineiro da rodovia federal BR 352. Hugo Aguiar nasceu na cidade mineira de Araguari, no dia 1º de março de 1928. Filho do Sr. José Emílio de Aguiar e Dona Carmen Porto Aguiar, estudou em sua terra natal até sair para a conclusão do secundário no Instituto Padre Machado, em Belo Horizonte. Ainda na capital das alterosas, cursou Direito na Universidade de Minas Gerais, bacharelando se em 1951. 2 Sua carreira política iniciou se com o mandato de vereador, em Coromandel, tendo também exercido, na Assembleia Legislativa do Estado de Minas Gerais, dois mandatos de Deputado Estadual, entre 1959 e 1963, na 4ª e 5ª legislaturas, pelo PSD, quando chegou a ocupar o cargo de vice presidente da Assembleia. Nesse mesmo período, chegou a licenciar se para ocupar o cargo de Secretário de Estado da Administração. Na Câmara Federal também foi por duas vezes Deputado, sempre pela ARENA, nas legislaturas 1967 1971 e 1971 1975, onde integrou a Comissão de Agricultura e Política, a Comissão de Finanças e a Comissão de Serviço Público. Além dos mandatos eletivos e do já citado cargo de Secretário de Estado de Administração, Hugo Aguiar também foi Chefe de Gabinete do Secretário de Interior e Justiça de Minas Gerais Diretor da Caixa Econômica do Estado de Minas Gerais MinasCaixa de 1975 a 1983 Delegado Federal do Ministério de Agricultura, de 1984 a 1986 e Presidente da Companhia Mineradora do Pirocloro de Araxá, entre 1988 e 1999. Destacado agricultor e advogado Hugo Aguiar soube exercer com dignidade todos os cargos públicos para os quais foi eleito ou designado, sempre buscando o bem estar dos cidadãos brasileiros, notadamente daqueles das Minas Gerais. Faleceu em Coromandel, no ano de 2003. O traçado mineiro da rodovia BR 352, objeto da denominação proposta, tem início na ponte sobre o rio Paranaíba divisa com o Estado de Goiás seguindo, entre outros, pelos Municípios de Abadia dos Dourados, Coromandel, Patos de Minas, Abaeté e Pitangui, até finalizar no entroncamento com a BR 262, em Pará de Minas. Em face do prestigioso currículo do ex Deputado Hugo Aguiar e dos seus valiosos serviços prestados ao Estado de Minas Gerais, esperamos contar com o apoio de nossos Pares para aprovarmos este projeto de lei, que atribui o seu nome à BR 352 em Minas Gerais. Sala das Sessões, em de de 2012. Deputado VITOR PENIDO 2011_19082 - PROJETO DE LEI Nº DE 2012 Do Sr. Mandetta Acrescenta inciso X ao art. 473 da Consolidação das Leis do Trabalho, a fim de permitir falta ao serviço do empregado para participar de trabalhos comunitários. O Congresso Nacional decreta Art. 1º O art. 473 da Consolidação das Leis do Trabalho, aprovada pelo Decreto lei n. º 5. 452, de 1º de maio de 1943, passa a vigorar acrescido do seguinte inciso X Art. 473 X por um dia, em cada doze meses de trabalho, para participar de trabalhos comunitários devidamente comprovados. NR Art. 2º Esta lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO Uma das principais formas de fomentar o engajamento social, a consciência de cidadania e o espírito de colaboração da sociedade com as políticas públicas de desenvolvimento é estimular a participação em movimentos comunitários. 2 Não é necessário detalhar os múltiplos benefícios sociais, relacionais e de qualidade de vida que surgem do envolvimento dos cidadãos com as iniciativas comunitárias. Os trabalhadores, pessoas economicamente ativas, precisam ser estimulados a aderirem a essas iniciativas. No entanto, as longas jornadas de trabalho durante a semana e os desafios da vida familiar pouco colaboram para o engajamento dessa significativa parcela da população nos trabalhos comunitários. Nada mais justo, em função do papel social da propriedade privada, que os empregadores liberem seus empregados, uma única vez a cada ano, para que estes experimentem e passem a aderir voluntariamente às diversas alternativas de engajamento comunitário. Seria criado, assim, um meio de estimular a adesão à programas de voluntariado no Brasil, onde os índices de participantes nesse tipo de projeto é tão baixo. Isso, visando criar, em um futuro próximo, a cultura de voluntariado nos brasileiros. Por estas razões, temos a certeza de contar com o apoio dos ilustres Deputados e Deputadas para a aprovação do presente projeto de lei. Sala das Sessões, em de de 2012. Deputado MANDETTA - PROJETO DE LEI Nº DE 2015 Do Sr. João Marcelo Souza Institui a Política Nacional para o Manejo Sustentável e Plantio da Palmeira do Babaçu Orbignya martiana e dá outras providências. O Congresso Nacional decreta Art. 1º Fica instituída a Política Nacional para o Manejo Sustentável e Plantio da Palmeira do Babaçu Orbidgnya martiana com as seguintes finalidades I desenvolver, financiar e modernizar a cultura da palmeira do babaçu II incentivar o aumento da produtividade do cultivo e exploração da palmeira do babaçu III estimular a produção de produtos derivados, o aproveitamento industrial, a exportação, a defesa de preços de comercialização e a abertura de mercados IV elevar a qualidade de vida dos trabalhadores do setor V criar uma rede de serviços de apoio para a cadeia do coco babaçu VI fomentar a criação e regulamentação de associações, cooperativas e centrais 2 VII promover o fomento e o crédito para a implantação de agroindústrias de base familiar para o aproveitamento integral do coco babaçu VIII definir e apoiar o desenvolvimento de manejo sustentável para o uso do babaçu. Art. 2º Fica proibida a derrubada e o uso predatório das palmeiras de babaçu existentes no território nacional, com exceção I. em áreas destinadas a obras e serviços de utilidade pública ou de interesse social declaradas pelo poder público II. com o objetivo de estimular a reprodução das espécies, aumentar a produção de seus frutos ou facilitar a sua coleta III. quando autorizado por órgão ambiental competente. Art. 3º Para a consecução dos objetivos previstos nesta Lei, a Política Nacional para o Manejo Sustentável e Plantio da Palmeira do Babaçu contará com os seguintes recursos I. dotações orçamentárias da União II. produto de operações de crédito internas e externas firmadas com entidades públicas, privadas, nacionais ou estrangeiras III. saldos de exercícios anteriores IV. outras fontes previstas em lei. Art. 4º Os recursos referidos no art 3º desta lei serão destinados a I apoiar o desenvolvimento da cultura da palmeira do babaçu, promovendo a disseminação de tecnologias que concorram para o aumento da sua produtividade e da qualidade do produto II fortalecer e expandir os segmentos da cadeia produtiva do babaçu III realizar pesquisas, estudos e diagnósticos 3 IV promover a capacitação tecnológica na indústria da cultura do babaçu e seu beneficiamento V realizar ampliações e melhorias na infraestrutura de apoio à produção e comercialização do babaçu e de seus derivados VI incrementar a cooperação técnica e financeira internacional com organismos particulares e oficiais, relativa à cultura do babaçu. Art. 5º Esta Lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO De acordo com a Embrapa Cocais, o Brasil dispõe de aproximadamente 18 milhões de hectares de babaçuais em seu território. O estado do Maranhão é o maior produtor nacional. Piauí, Pará, Mato Grosso e Tocantins também dispõem de babaçuais. O método de beneficiamento do babaçu é bastante rudimentar e dependente das quebradeiras de coco que são as mulheres que executam o trabalho manual da colheita e extração da amêndoa, em condições frequentemente precárias. A palmeira do babaçu chega a atingir 20 metros de altura e pode ser totalmente aproveitada. Do babaçu se obtém a matéria prima usada na fabricação de cosméticos, sabão, banha de coco e margarina. Do seu broto se extrai o palmito. Quando maduro, a parte exterior é comestível. O caule é utilizado em construções na área rural. As suas folhas são usadas para cobertura de casas e abrigos. Serve, também, para fabricação de papel e celulose. Do pedúnculo extrai se um líquido que, quando fermentado, transforma se em bebida alcóolica. O babaçu serve também para a fabricação de cestos, esteiras, janelas, gaiolas, entre outros. De acordo com o sítio PRISMABRASIL, as amêndoas verdes recém extraídas e espremidas com um pouco de água em um pano fino fornecem um leite de propriedades nutritivas semelhantes às do leite humano, 4 segundo pesquisa do Instituto de Recursos Naturais do Maranhão. Esse leite é muito usado na culinária local como tempero para carnes de caça e peixes substituindo o leite de coco da baía e como mistura para empapar o cuscuz de milho, de arroz e de farinha de mandioca ou, até mesmo, bebendo ao natural, substituindo o leite de vaca. A casca de coco, devidamente preparada, fornece um eficiente carvão, fonte exclusiva de combustível em várias regiões do nordeste do Brasil. Outros produtos de aplicação industrial podem ser derivados da casca de coco de babaçu, tais como etanol, metanol, coque, carvão reativado, gases combustíveis, ácido acético e alcatrão. É, portanto, um dos principais produtos extrativos do Brasil, contribuindo sobremaneira para a economia de alguns estados da Federação. O babaçu desempenha, também, relevante papel social, principalmente pela grande capacidade de absorção de mão de obra, especialmente na entressafra das culturas tradicionais. De acordo com o Projeto Executivo resultante do Congresso Brasileiro de Palmeira de Babaçu, realizado em 2010, em São Luís do Maranhão, nos últimos anos, o babaçu passou a assumir crescente grau de importância, dentre os recursos renováveis da biomassa do país, como fonte alternativa de geração de energia. No Modelo Energético Nacional o babaçu, é classificado como uma alternativa mais promissora. Para o Estado do Maranhão, Piauí, Tocantins, Mato Grosso e Goiás servirão como participante ativo em alto grau de complementaridade, através de seu consorciamento e balanceamento térmico na produção de combustível diesel e lubrificante. A despeito de tantas utilidades e do grande potencial de geração de renda para inúmeras famílias, o babaçu continua a ser tratado de forma marginal, permanecendo, somente, como parcela integrante dos sistemas de subsistência. São por essas razões que apresentamos o presente Projeto de Lei, vez que, intenta apoiar o desenvolvimento da cultura da palmeira de babaçu, promovendo a disseminação de tecnologias que concorram para o aumento da sua produtividade e da qualidade do produto. Por oportuno, salientamos que, para apresentação da presente proposição, nos inspiramos no Projeto de Lei nº 6. 820 de 2013, que Institui o fundo Nacional de Apoio à cultura da Palmeira do Babaçu, do Deputado Costa Ferreira, arquivado nos termos do art. 105 do Regimento 5 Interno da Câmara dos Deputados, bem como em levantamentos do Plano Nacional para a Promoção dos Produtos da Sociobiodiversidade, criado em 2009. Contamos, assim, com o apoio de nossos nobres pares, no sentido de aperfeiçoar e aprovar o texto proposto. Sala das Sessões, em de de 2015. Deputado JOÃO MARCELO SOUZA PMDB MA - source_sentence: PROJETO DE LEI N ____________ 2007 Altera a redação do art. 23 da Lei n 8. 666, de 21 de junho de 1993, e dá outras providências. A Congresso Nacional decreta Art. 1 Os incisos I e II do art. 23 da Lei n 8. 666, de 21 de junho de 1993, passam a vigorar com a seguinte redação Art. 23. As modalidades de licitação a que se referem os incisos I a III do artigo anterior serão determinadas em função dos seguintes limites, tendo em vista o valor estimado da contratação I para obras e serviços de engenharia a convite até R 325. 973, 00 trezentos de vinte e cinco e novecentos e setenta e dois reais b tomada de preços até 3. 259. 726, 00 três milhões, duzentos e cinqüenta e nove mil e setecentos e vinte e cinco reais c concorrência acima de 3. 259. 726, 00 três milhões, duzentos e cinqüenta e nove mil e setecentos e vinte e cinco reais II para compras e serviços não referidos no inciso anterior a convite até R 132. 054, 00 cento e trinta e dois mil e cinqüenta e quatro reais b tomada de preços até R 1. 072. 938, 00 um milhão, setenta e dois mil e novecentos e trinta e oito reais c concorrência acima de 1. 072. 938, 00 um milhão, setenta e dois mil e novecentos e trinta e oito reais Art. 2 Esta lei entra em vigor na data de sua publicação. Art. 3 Ficam revogadas as disposições em contrário. JUSTIFICATIVA A lei n 8. 666, de 21 de junho de 1993, regulamenta as modalidades de licitações, estabelecendo limites, de acordo com o valor da contratação, para o uso de cada modalidade. Tais limites, constantes no art. 23 da referida lei, foram atualizados pela última vez em 1998, com o advento da Lei n 9. 648, o que gerou uma enorme defasagem dos valores constantes nesse dispositivo. Desta forma, busca a pretensa lei atualizar os valores que limitam as licitações de 1998 até ao presente ano. Os valores descritos foram atualizados de acordo com o INCC DI, da Fundação Getúlio Vargas, no caso do inciso I, e com o IGP DI, também da Fundação Getúlio Vargas. Diante do exposto, justifica se a presente proposta de lei. Brasília, 10 de agosto de 2007. MANOEL ALVES DA SILVA JUNIOR Deputado Federal sentences: - CÂMARA DOS DEPUTADOS PROJETO DE LEI Nº______, DE 2013 Do Sr. Davi Alves Silva Júnior Obriga as unidades do Sistema Único de Saúde a promoverem a marcação de procedimentos de atenção à saúde pelos usuários, por meio da rede mundial de computadores. O Congresso Nacional decreta Art. 1º Esta Lei obriga as unidades do Sistema Único de Saúde SUS a promoverem a marcação de procedimentos de atenção à saúde pelos usuários, por meio da rede mundial de computadores. Art. 2º Ficam as instâncias gestoras do Sistema Único de Saúde, em todas as esferas de governo, obrigadas a promover a marcação de procedimentos de atenção à saúde pelos usuários, por meio da rede mundial de computadores, nas unidades de saúde sob sua responsabilidade direta. Parágrafo único. As instâncias gestoras mencionadas no caput deste artigo ficam obrigadas a disponibilizar terminais de acesso, preferencialmente, dentro das unidades de saúde ou em centros de atendimento em locais próximos das unidades, para que o usuário efetive a marcação de procedimentos. Art. 3º A obrigação referida no artigo 2º será aplicada para qualquer procedimento de atenção à saúde em que seja necessária a marcação de horário para atendimento em unidade de saúde do SUS. Art. 4º Esta Lei entra em vigor doze meses após a data de sua publicação. CÂMARA DOS DEPUTADOS 2 JUSTIFICAÇÃO Uma das situações mais degradantes que o cidadão brasileiro enfrenta para exercer seu direito constitucional à saúde por meio do Sistema Único de Saúde SUS é o enfrentamento de longas filas de espera para efetuar a marcação de procedimentos de atenção à saúde, como consultas e exames. Muitas vezes, precisa sair de sua casa de madrugada, passando a noite e, até, o dia, esperando pela marcação. Essa proposição objetiva contribuir para a solução desse problema ao prever a obrigação de que as instâncias gestoras do SUS, em todas as esferas de governo, promovam a marcação de procedimentos por meio da rede mundial de computadores, nas unidades de saúde sob sua responsabilidade direta. Tal medida resultaria em maior celeridade para o usuário do SUS e também para as próprias unidades de saúde. Foi previsto que as instâncias gestoras devem disponibilizar terminais de acesso, preferencialmente dentro das unidades de saúde, para que o usuário efetive a marcação. O ideal seria que toda unidade de saúde oferecesse acesso à marcação pela Internet aos seus usuários, contudo a realidade do País ainda não permite isso, pois nem todas possuem acesso à Internet. Ainda assim, nesses casos o acesso será promovido em local próximo à unidade de saúde centro de atendimento que possua tal acesso. A obrigação abrangerá qualquer procedimento de atenção à saúde em que seja necessária a marcação de horário para atendimento em unidade de saúde do SUS. Desse modo, ficam incluídos os procedimentos que são eletivos, pois os de urgência e emergência devem ser atendidos de imediato, sem a necessidade de marcação. A proposição indica que a Lei entrará em vigor doze meses após a data de sua publicação, para que haja tempo hábil para sua implantação. CÂMARA DOS DEPUTADOS 3 Diante da relevância da matéria para melhorar a qualidade da atenção no SUS, solicito o apoio dos ilustres Pares para aprovar a proposição nesta Casa. Sala das Sessões, em de de 2013. Deputado DAVI ALVES SILVA JÚNIOR 2013_4984 - Altera a Lei nº 8. 666, de 21 de junho de 1993, para reajustar os valores estimados de contratação pela Administração Pública. O Congresso Nacional decreta Art. 1º O art. 23 da Lei nº 8. 666, de 21 de junho de 1993, passa a vigorar com as seguintes alterações Art. 23. As modalidades de licitação a que se referem os incisos I a III do art. 22 serão determinadas em função dos seguintes limites, tendo em vista o valor estimado da contratação I para obras e serviços de engenharia a convite até R 225. 000, 00 duzentos e vinte e cinco mil reais b tomada de preços até R 2. 250. 000, 00 dois milhões, duzentos e cinqüenta mil reais c concorrência acima de R 2. 250. 000, 00 dois milhões, duzentos e cinqüenta mil reais II para compras e serviços não referidos no inciso I a convite até R 120. 000, 00 cento e vinte mil reais b tomada de preços até R 975. 000, 00 novecentos e setenta e cinco mil reais c concorrência acima de R 975. 000, 00 novecentos e setenta e cinco mil reais NR Art. 2º Esta Lei entra em vigor na data de sua publicação. Senado Federal, em 11 de junho de 2002 Senador Ramez Tebet Presidente do Senado Federal vpl pls02 061 - PROJETO DE LEI No DE 2006 Do Sr. Luiz Carlos Hauly Altera os arts. 1º, caput 5º, I 6º, 4º e 7º 49 52, 4º e 71, incisos I e II, bem como revoga os arts. 57 e 68 da Lei nº 11. 101, de 9 de fevereiro de 2005, para eliminar a correção monetária do plano especial de recuperação judicial para microempresas e empresas de pequeno porte, e dá outras providências. O Congresso Nacional decreta Art. 1º Os arts. 1º, caput 5º, I 6º, 4º e 7º 49 52, 4º e art. 71, incisos I e II, da Lei nº 11. 101, de 9 de fevereiro de 2005, passam a vigorar com a seguinte redação Art. 1º Esta Lei disciplina a recuperação judicial, a falência e a recuperação extrajudicial do empresário, da sociedade cooperativa e da sociedade empresária, doravante referidos simplesmente como devedor. Art. 5º I as obrigações a título gratuito, aval e fiança Art. 6º 4º Na recuperação judicial, a suspensão de que trata o caput deste artigo dar se á pelo prazo de 180 cento e oitenta dias, contados do deferimento do processamento da recuperação, prorrogável uma única vez por até 90 noventa dias, a critério do juízo ou caso a assembléia geral de credores não tenha aprovado o respectivo plano de recuperação judicial e, após o decurso desse prazo, será restabelecido o direito dos 2 credores de iniciar ou continuar suas ações e execuções, independentemente de pronunciamento judicial. 7º As execuções de natureza fiscal ficam suspensas a partir do deferimento da recuperação judicial. Art. 49. Estão sujeitos à recuperação judicial todos os créditos existentes na data do pedido, ainda que não vencidos, inclusive os créditos tributários. Art. 52. 4º O devedor não poderá desistir do pedido de recuperação judicial após o deferimento de seu processamento, salvo se obtiver aprovação da desistência na assembléia geral de credores, que obtenha, no mínimo, quorum de 50 cinqüenta porcento dos credores presentes, ficando impossibilitado de renovar pedido de recuperação pelo período de 2 dois anos, contados da homologação da desistência pelo juízo. Art. 71........................................................................ I abrangerá todos os credores II preverá o parcelamento em até 48 quarenta e oito parcelas mensais iguais e sucessivas, acrescidas de juros de 6 seis por cento ao ano III IV Parágrafo único. NR Art. 2º Ficam revogados os arts. 57 e 68 da Lei nº 11. 101, de 9 de fevereiro de 2005. Art. 3º Esta Lei entra em vigor na data de sua publicação oficial. 3 JUSTIFICAÇÃO A Lei nº 11. 101 que regulamenta a recuperação judicial, a extrajudicial e a falência do devedor empresário ou da sociedade empresária, que entrou em vigor no mês de junho de 2005, contém algumas impropriedades que, a nosso ver, carecem de um urgente aperfeiçoamento nesta Casa. Primeiramente, queremos destacar a problemática dos créditos fiscais que vêm dificultando, em muito, o acesso das empresas ao plano de recuperação judicial, levando as em direção à pior alternativa legal, que é a falência. Ocorre que a nova lei, em seu art. 6º, 7º, determinou a não suspensão das execuções de natureza fiscal com o deferimento da recuperação judicial. Tal medida, na prática, excluiu os créditos fiscais do processo de recuperação, a despeito de, no mesmo parágrafo 7º, o Legislador ter ressalvado a possibilidade de o empresário pleitear o parcelamento desses créditos junto à autoridade tributária, observados os termos do Código Tributário Nacional e da legislação ordinária específica. Tal tratamento aos créditos fiscais na recuperação judicial nos parece muito incoerente e inibe o acesso das empresas com passivo fiscal ao instituto da recuperação judicial. Entendemos que os créditos fiscais devem, sim, ser submetidos à recuperação judicial, como, aliás, defendemos durante a fase de discussão da proposição na sua primeira tramitação nesta Casa. Do mesmo modo, também estamos propondo a supressão dos art. 57 e 68. Na prática, em alguns casos já verificados após o advento da nova lei, configura se como inviável, para as empresas em vias de recuperação, cumprir a exigência contida no art. 57 da nova lei, qual seja a de apresentar a prova de quitação de débitos para com o Fisco, como pré requisito para terem o pedido de recuperação deferido pelo Juiz. Ora, uma vez que a empresa se encontra em difícil situação econômico financeira, e projeta se à elaboração de um plano de recuperação judicial, é porque não está conseguindo assumir as dívidas pactuadas com os credores, e, conseqüentemente, com o Fisco. As empresas que se encontram em tal situação, com um expressivo passivo fiscal, dificilmente conseguirão arcar com os seus débitos 4 fiscais, mostrando se, portanto, inviável o atendimento da exigência da prova de quitação dos débitos perante o Fisco. Isto ocorre porque a empresa em dificuldades prioriza, como é o lógico e o mais pertinente a ser feito, o destino de seus poucos proventos que ainda restam ao pagamento de seus empregados e fornecedores. Esta prioridade dá se em virtude da necessidade de manter a empresa em funcionamento. Todos sabemos que uma empresa que não paga os seus funcionários, quanto menos seus fornecedores, jamais terá condições de continuar com as suas atividades. Por isto, é elementar a prioridade do deslocamento das verbas restantes, em desfavor do Fisco. Saliente se que a certidão negativa somente poderá ser recusada se houver crédito efetivamente constituído contra o interessado. O simples descumprimento de obrigação tributária, principal ou acessória, não enseja motivo suficiente para a autoridade competente recusar a expedição, a menos que já tenha sido feito o lançamento do crédito correspondente. Ademais, como o objetivo primordial da nova lei ao buscar a superação de crise econômico financeira do devedor é o de assegurar a manutenção da fonte produtora de bens e serviços e dos empregos, consideramos ser imprescindível incluir no art. 1º, que estabelece o alcance da nova lei, as sociedades cooperativas, visto que essas entidades estão atravessando enormes dificuldades econômicas e financeiras sobre as quais os novos dispositivos da LRJ F não pode trazer qualquer alívio ou solução jurídica. Certamente, a nova lei se constitui numa grande conquista para a modernização de nossa economia e, especialmente, para a reestruturação de nossas empresas, que vêm lutando há anos contra uma excessiva carga tributária e contra os escorchantes juros cobrados em nosso País. Deixar as cooperativas à margem da nova lei constitui se num equívoco que poderá trazer sérios prejuízos a milhares de brasileiros que operam num setor muito expressivo da economia nacional. De outro modo, a nosso ver, a Lei nº 11. 101 05 foi muito dura com o tratamento dado às microempresas e empresas de pequeno porte, fugindo, inclusive, do mandamento constitucional contido no inciso IX do art. 5 170, que determina tratamento favorecido para as empresas de pequeno porte constituídas sob as leis brasileiras e que tenham sua sede e administração no país Assim, também propomos uma urgente e necessária modificação nos termos do plano de recuperação judicial que deverá ser apresentado por microempresa ou empresa de pequeno porte, uma vez que a atual redação dos incisos I e II do art. 71 da Lei nº 11. 101 05 se configura absolutamente prejudicial àquelas empresas, trazendo uma injustificada discriminação e excessivo ônus para o processo de recuperação judicial das empresas inseridas nesse importante segmento da economia nacional. A atual redação do inciso I do art. 71 determina que o plano de recuperação judicial das microempresas e empresas de pequeno porte abrangerá exclusivamente os créditos quirografários, deixando de fora os credores que forem bancos e outras instituições integrantes do Sistema Financeiro Nacional. Ora, como uma microempresa que deve a bancos poderá efetivamente pleitear sua recuperação, se os créditos detidos por bancos estão excluídos do plano Trata se certamente de um contra senso e de um absurdo que precisa ser corrigido urgentemente. Ademais, o inciso II do mesmo artigo estipula um parcelamento em apenas 36 meses e correção monetária e juros de 12 para o pagamento das parcelas devidas pelas microempresas e empresas de pequeno porte submetidas a processo de recuperação judicial. Mais uma vez, a nova lei não se coadunou com o art. 170, IX, da Constituição Federal, na medida em que não propiciou um prazo factível e suficiente para a reorganização financeira das microempresas e empresas de pequeno porte, como tampouco lhes amenizou os encargos sobre o passivo devido. Tal exigência, aliás, não existe na mesma lei para empresas de médio e grande porte, uma vez que o plano de recuperação dessas empresas poderá prever abatimento de juros, total ou parcial, desde que haja anuência dos seus credores. 6 Por estas razões, contamos com o apoio de nossos Pares para incluir estas modificações na Lei nº 11. 101 05, objetivando possibilitar condições mais justas e equânimes no processo de recuperação judicial para as sociedades cooperativas e microempresas e empresas de pequeno porte, que são tão importantes para a economia nacional. Sala das Sessões, em de de 2006. Deputado LUIZ CARLOS HAULY PSDB PR - source_sentence: PROJETO DE LEI Nº DE 2013 Do Sr. Vitor Penido Altera o art. 2º da Lei nº 11. 096, de 13 de Janeiro de 2005 que institui o Programa Universidade Para Todos PROUNI. O Congresso Nacional decreta Art. 1º O inciso I do art. 2º da Lei nº 11. 096, de 13 de Janeiro de 2005 fica com a seguinte redação Art. 2º A bolsa será destinada a I Estudante que tenha cursado o ensino médio integralmente em escola pública ou que comprove renda familiar que impeça o pagamento da mensalidade escolar em instituição de ensino superior. Art. 2º. Esta lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO O Programa Universidade Para Todos representou uma importante contribuição para a democratização do acesso ao ensino superior. Exatamente por sua relevância merece medidas que o aperfeiçoem. 2 O inciso I do art. 3º da Lei nº 11. 096 de 2005, que institui o Programa Universidade Para Todos, restringe o acesso às bolsas que oferece apenas a candidatos que tenham cursado todas as séries do ensino médio em escola pública ou que as tenham cursado em escola particular na condição de bolsista integral. A premissa que orienta o dispositivo é a de que as famílias dos alunos nas situações nele estipuladas não contariam com as condições financeiras para arcar com os custos das mensalidades nos estabelecimentos de ensino superior. O princípio está correto apenas parcialmente, pois há que se considerar o caso de estudantes cujas famílias tiveram queda acentuada de renda durante ou após a conclusão do ensino médio. Esta é uma possibilidade real em vista da crise econômica e do desemprego que assola a classe média brasileira os estratos de renda mais elevada da classe média têm visto seus números decrescerem dramaticamente nas últimas décadas. É comum que, perdendo o emprego o chefe de família, seus filhos tenham que sair da escola particular. Porém, o fato de terem por um período, por pequeno que seja, pago as mensalidades de um estabelecimento privado de ensino desqualifiqua os para a candidatura ao Prouni. Há, ainda, o caso de famílias que nunca fizeram parte a da classe média mas que em extremo sacrifício conseguiram pagar um estabelecimento particular durante o ensino médio. E também não se qualificam para o PROUNI pelo critério atual. Trata se de evidente injustiça que este projeto de lei vem corrigir. Sala das Sessões, em de de 2013.. Deputado VITOR PENIDO 2007_1624_Vitor Penido_145 sentences: - Projeto de Lei n. º____, de 2008. Da Deputada Luciana Genro PSOL RS Dá nova redação ao inciso VI, do art. 46 da lei nº 9. 610, de 19 de fevereiro de 1998, que Altera, atualiza e consolida a legislação sobre direitos autorais e dá outras providências. O Congresso Nacional Decreta Art. 1º Esta Lei dá nova redação ao inciso VI, do art. 46 da Lei n. º 9. 610, de 19 de fevereiro de 1988, que altera, atualiza e consolida a legislação sobre direitos autorais e dá outras providências. Art. 2º O inciso VI, do art. 46 da Lei n. º 9. 610, de 19 de fevereiro de 1988, passa a vigorar com a seguinte redação Art. 46. Não constitui ofensa aos direitos autorais VI a representação teatral e a execução musical realizadas no recesso familiar, nos estabelecimentos de ensino e naqueles destinados à prática desportiva NR JUSTIFICAÇÃO A Constituição Federal art. 5º, inciso XXVII assegura o direito do autor e a Lei n. º 9. 610, de 1998, regula a matéria relativa aos direitos autorais. O exercício do direito de autor implica, nos termos da Lei 9. 610, na cobrança pelo Escritório Central de Arrecadação ECAD, de valor pecuniário, por exemplo, pela execução musical com intuito de lucro. Apesar de tratar se cláusula pétrea e direito individual, o direito de autor, principalmente numa república democrática, deve ser entendido conjugado com outros direitos e deveres, dentre eles o direito à cultura, educação, desporto e comunicação e o dever da função social da propriedade. 1 C 2 7 F B E 1 3 2 1 C 27 FB E 13 2 Esta função social, aliás, é igualmente um poder dever pétreo e individual homogêneo, vez que consta do mesmo rol do mencionado art. 5º da CF inciso XXIII A atual redação do inciso VI do art. 46 da lei de direitos autorais tem se mostrado prejudicial ao direito constitucional à privacidade, aos direitos do consumidor, ao aceso ao conhecimento, ao acesso à cultura. Tem contribuído para uma interpretação judicial restritiva da lei de direito autoral no Brasil, nas milhares de ações judiciais que intentam a cobrança pela execução musical em estabelecimentos de prática desportiva. Vemos que, como já foi dito, pela mera referência à reprodução eventual de alguma música de autores nacionais não pode dar amparo à cobrança dos direitos autorais. É certo, ainda, que a finalidade de execução musical em estabelecimentos onde há prática desportiva não é essencialmente comercial para o destinatário da reprodução o desportista e que o efeito da execução não acarreta prejuízo à exploração normal da obra pelo autor, mas, ao contrário, pode até estimulá la. Estas são as razões pelas quais apresentamos o presente projeto, rogando pelo apoiamento dos pares. Sala das Sessões, 11 de junho de 2008. Deputada Luciana Genro PSOL RS 1 C 2 7 F B E 1 3 2 1 C 27 FB E 13 2 - CÂMARA DOS DEPUTADOS PROJETO DE LEI Nº DE 2015 Do Sr. Takayama Dispõe sobre a prestação de assistência religiosa e espiritual por meio de capelania nos estabelecimentos que menciona e dá outras providencias. O Congresso Nacional decreta Art. 1º Fica assegurada a assistência religiosa e espiritual por meio do serviço de capelania em hospitais da rede pública ou privada, em estabelecimentos prisionais civis ou militares, estabelecimentos de ensino, entidades sócio educativas, bem como quartéis no âmbito do nosso país. 1º Entende se por serviço de capelania, entre outros, os seguintes I aconselhamento II orientações aos assistidos III cultos e orações IV ministrar a Santa Comunhão V ministrar a palavra. 2º A assistência religiosa e espiritual de que trata o caput será ministrada por Capelão devidamente constituído. 3º Os estabelecimentos a que se refere o caput deste artigo manterão local apropriado para os cultos religiosos. Art. 2º São beneficiários da assistência de que trata esta lei I discentes e docentes dos estabelecimentos de ensino da rede pública ou privada II pacientes internados em hospitais públicos e privados CÂMARA DOS DEPUTADOS III reclusos em estabelecimentos prisionais, delegacias, quartéis, ou estabelecimentos sócio educativos IV militares no ambiente dos quartéis. Parágrafo único Somente poderá ser prestada a assistência religiosa a que se refere esta Lei mediante manifestação dos interessados, uma vez que nenhum assistido poderá ser obrigado a participar das atividades religiosas. Art. 3º As instituições religiosas que desejarem prestar a assistência de que trata esta lei, deverão cadastrar se preferencialmente no Instituto Nacional da Justiça de Paz e Juízes de Paz do Brasil INJUPA Br, e ou em qualquer instituição credenciadora que ministre o curso de capelania. Parágrafo único A instituição credenciadora deverá ser legalmente constituída, obedecidos os requisitos e limites de atuação estabelecidos pela legislação vigente, mediante a apresentação de documento contendo os atos constituídos, devidamente registrado junto a uma ordem regulamentadora da atividade. Art. 4º O interessado em obter a credencial para exercer a atividade de que trata esta Lei deverá apresentar o termo de apresentação, identificação, idoneidade e responsabilidade, subscrito pelo INJUPA Br ou pela instituição credenciadora a qual pertença. Art. 5º Será criado e mantido pelo INJUPA Br um registro de identificação de pessoas credenciadas, na sua instituição. Art. 6º O cartão de credenciamento conterá, além da identificação pessoal, foto recente do credenciado e sua validade limita se a 1 um ano. Art. 7º São requisitos indispensáveis para o credenciamento dos interessados I ser maior de 21 anos II estar no exercício de seus direitos políticos, se brasileiro III estar em condição regular no país, se estrangeiro IV ser pessoas de ilibada conduta moral e profissional V ser apresentado por entidade religiosa interessada, nos termos do art. 10 desta Lei CÂMARA DOS DEPUTADOS VI ser habilitado por instituição de capelania e registrado em entidade reconhecida como regulamentadora da atividade, tendo cumprido as exigências impostas pela lei em vigor. Art. 8º Para os fins da aplicação do disposto nesta lei, fica garantida a livre prática de culto para todas as crenças religiosas aos assistidos e seus familiares, sendo permitindo a participação nos serviços organizados nos estabelecimentos a que se refere o art. 1º desta lei, tendo em vista o interesse prevalecente da coletividade. Parágrafo Único Os Capelães poderão prestar concursos públicos ou ser contratados na Marinha, Exército, Aeronáutica, hospitais, presídios, Instituições Militares, instituições carcerárias e respectivas entidades sócio educativas, desde que atendam os requisitos exigidos pelos mesmos. Art. 9º Os Capelães de instituições legalmente constituídas, quando apresentados por estas, poderão ser supervisionados por outro Capelão quando forem prestar serviços auxiliares de assistência religiosa e espiritual. Art. 10º Será garantido o acesso de Capelães, desde que devidamente credenciados nos termos desta Lei, às dependências das unidades hospitalares, prisionais e sócio educativas, bem como dos estabelecimentos de ensino, com a finalidade de assistência religiosa e espiritual, ficando dispensados, no caso dos estabelecimentos prisionais, da revista manual, na visita assistida, mediante a colaboração e segurança dos agentes penitenciários. 1º A credencial a que se refere o caput será emitida pelo Instituto Nacional da Justiça de Paz e dos Juízes de Paz do Brasil INJUPA Br, ou pela instituição credenciadora a qual pertença. 2º A assistência prestada pelos capelães inclui o sigilo no caso de entrevistas com presos e de confidências destes, de internados e funcionários. Art. 11 As instituições cadastradas poderão requerer credenciamento especial para que o capelão tenha acesso livre, ou seja, possa ingressar, visitar, e permanecer em qualquer dependência dos estabelecimentos a que se refere o art. 1º desta Lei. Art. 12 Os locais e horários para realização das atividades e cerimônias religiosas serão definidos pela direção dos estabelecimentos citados nesta Lei, podendo a assistência religiosa e espiritual, que poderá ser prestada fora dos horários normais de visitas, sendo que os Capelães deverão contar com a colaboração necessária ao desempenho de suas atribuições. CÂMARA DOS DEPUTADOS Art. 13 O descumprimento desta Lei, quanto às faculdades e garantias da pessoa credenciada, gera responsabilidade disciplinar imputável ao agente público que lhe der causa. Art. 14 Esta Lei entra em vigor na data de sua publicação. Brasília, 10 de agosto de 2015 Deputado HIDEKAZU TAKAYAMA JUSTIFICAÇÃO A atividade de capelania pressupõe uma abordagem do ser humano como criatura de Deus que apresenta potencialidades e necessidades físicas, intelectuais, emocionais e espirituais. Por exemplo, um Capelão integrante da equipe multidisciplinar de saúde, é uma pessoa capacitada e sensível às necessidades humanas, dispondo se a dar ouvidos, confortar e encorajar, ajudando o enfermo a lutar pela vida, com esperança em Deus e na medicina. Oferece aconselhamento espiritual e apoio emocional tanto ao paciente e seus familiares, como aos profissionais da saúde. E um importante elo com a comunidade local. Assim os Capelães são homens e mulheres preparados para resgatar vidas, levando aos assistidos palestras e seminários acerca de como viver uma vida melhor, embora estejam no ambiente de hospitais, clinicas e presídios. Os formandos para o curso de capelania, independentemente da faixa etária, são treinados para resgatar pessoas do sofrimento, e estão capacitados para trabalhar na prevenção da violência, do uso das drogas, contra pedofilia, na recuperação e na reabilitação de drogados, recuperação de meninos de rua, mendigos e presidiários, promovendo, com sua atividade, a cultura de paz aos assistidos e suas famílias por meio das visitas em hospitais, escolas e presídios. A expectativa é de que essa atividade de capelania viabilize mudanças fundamentais e comportamentais no seio da sociedade, com as pessoas se tornando mais conscientes dos seus deveres humanitários e mais solidárias para com aqueles que vivem em situação de risco e de vulnerabilidade devido à violência e ao uso de drogas, e, dessa forma, se sintam motivadas e encorajadas a combater e irradiar os malefícios do uso e do abuso de drogas. CÂMARA DOS DEPUTADOS Vale destacar que os males causados pelas drogas tem impactado sobremaneira a vida dos cidadãos de bem e da família brasileira, ensejando atos de violência e causando danos significativos na saúde física e emocional dos usuários. Enfim, esperamos alcançar as metas propostas e contribuir decisivamente para melhoria da qualidade de vida de pessoas das mais diversas faixas etárias que pretendemos assistir nas comunidades de todo o nosso país. Nesse sentido, é essencial que o Parlamento e a Administração Pública não se omitam quanto a esse papel de oferecer aos assistidos e a suas famílias um serviço fundamentado na manifestação de altruísmo, amizade, fraternidade, capaz de promover a Paz e a Solidariedade cidadã. É o que pretendo com a apresentação dessa proposição. Brasília, 10 de agosto de 2015 Deputado HIDEKAZU TAKAYAMA - Altera o art. 2º da Lei nº 11. 096, de 13 de janeiro de 2005, para estender o atendimento do Programa Universidade para Todos aos estudantes beneficiados com bolsa parcial no ensino médio privado. O Congresso Nacional decreta Art. 1º O art. 2º da Lei nº 11. 096, de 13 de janeiro de 2005, passa a vigorar acrescido do seguinte 1º, renumerando se como 2º o atual parágrafo único Art. 2º 1º Após a distribuição prevista no caput deste artigo, eventual bolsa excedente poderá ser destinada a estudante que tenha cursado o ensino médio, ou parte dele, em escola da rede privada, na condição de bolsista parcial, nos termos definidos em regulamento do Ministério da Educação, observada a devida proporcionalidade com o tempo ou percentual de estudos gratuitos. 2º antigo parágrafo único NR Art. 2º Esta Lei entra em vigor na data de sua publicação. Senado Federal, em de dezembro de 2006. Senador Renan Calheiros Presidente do Senado Federal acf pls05 260 - source_sentence: PROJETO DE LEI Nº DE 2015 Do Sr. Veneziano Vital do Rêgo Revoga dispositivos da Lei nº 12. 462, de 4 de agosto de 2011, para limitar a abrangência do Regime Diferenciado de Contratações aos casos inicialmente previstos pela lei original, e dá outras providências. O Congresso Nacional decreta Art. 1º Revogam se os incisos IV, V e VI, e o 3º do artigo 1º da Lei nº 12. 462, de 4 de agosto de 2011. Art. 2º Revogam se os artigos 1º e 2º da Lei nº 12. 873, de 24 de outubro de 2013. Art. 3º Revoga se o art. 15 A da Lei nº 12. 340, de 1º de dezembro de 2010. Art. 4º Esta Lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO A Lei nº 12. 462, de 4 de agosto de 2011, resultante da conversão em lei da Medida Provisória nº 527, de 2011, instituiu o Regime Diferenciado de Contratações Públicas RDC, regime criado para atender às obras para a Copa do Mundo e da Olimpíada. Ou seja, a diferenciação e as 2 facilidades previstas na referida lei foram previstas, excepcionalmente, para viabilizar as obras e serviços necessários para a realização dos grandes eventos em nosso País. Ressalta se que os objetivos iniciais acabaram sendo maculados, uma vez que foram inseridos no corpo da lei, e em outras leis, autorizações para a utilização permanente do RDC para finalidades bem distintas das originalmente previstas. Portanto, faltando pouco mais de um ano para a realização das Olimpíadas no Estado do Rio de Janeiro, entendemos que ao término do evento a Lei terá cumprido com os seus propósitos e não subsistirá razão para que qualquer outra obra venha a ser executada utilizando se do RDC como instrumento de licitação, voltando se a adotar integralmente, em toda a Administração Pública, os parâmetros da Lei nº 8. 666, de 21 de junho de 1993. Isto posto, solicitamos o apoio dos nobres colegas nesta Casa para aprovação do presente projeto de lei. Sala das Sessões, em de de 2015. Deputado VENEZIANO VITAL DO REGO 2015 9374 sentences: - PROJETO DE LEI Nº DE 2011 Do Sr. Laércio Oliveira Altera a Lei nº 11. 901, de 12 de janeiro de 2009. O Congresso Nacional decreta Art. 1º A presente lei é editada visando incluir na norma competente a obrigatoriedade de contratação de bombeiro civil em edificações ou conjunto de edificações com fins comerciais. Art. 2º A Lei nº 11. 901, de 12 de janeiro de 2009, passa a vigorar acrescida do seguinte artigo Art. 10 A Toda e qualquer edificação ou conjunto de edificações públicas e particulares, comerciais ou assemelhadas, tem a obrigação de contratar bombeiro civil, devidamente qualificado, para fazer parte do quadro permanente de pessoal. Parágrafo único. Tal contratação poderá ser realizada, conforme o art. 2º da presente lei, por intermédio de empresa especializada em prestação de serviços de prevenção e combate a incêndio. NR Art. 3º Ficam revogadas todas as disposições que contrariam o que fora tratado na presente norma. Art. 4º Esta lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO Com a publicação da presente lei, acidentes como o ocorrido no município do Rio de Janeiro serão prevenidos. Isso porque, com a atuação permanente de um bombeiro civil, situações de perigo podem ser antecipadas e ações de evacuação de edificações comerciais em iminente risco de incêndio ou explosão ocorrerão de forma correta e prudente. O caso concreto citado foi provocado por desrespeito às normas de segurança e, também, por incompetência da entidade publica que emitiu alvará de funcionamento sem atentar às especificidades do edifício. Mas, mesmo assim, não podemos ignorar o fato de que se houvesse um profissional devidamente habilitado trabalhando permanentemente no quadro de pessoal daquele prédio as vítimas poderiam ter sido salvas. Este indivíduo teria a discernimento técnico suficiente para avaliar a situação e promover a imediata evacuação das pessoas ali presentes, procedendo às devidas cautelas contra o pânico. A fatalidade aconteceu porque nenhum dos cidadãos foi capaz de determinar que o vazamento de gás colocasse suas vidas em extremo risco. Sendo assim, com a publicação da presente lei, situações como essas poderão ser extintas. Portanto, nesse contexto, rogo aos meus pares apoio na aprovação integral da presente proposição. Sala das Sessões, em de de 2011. LAÉRCIO OLIVEIRA Deputado Federal PR SE - Projeto de Lei nº 2011 Do Sr. Alessandro Molon PT RJ Dá nova redação ao art. 132 da Lei nº 8. 069, de 13 de julho de 1990, que Dispõe sobre o Estatuto da Criança e do Adolescente acrescentando parágrafos para definir quantidade e distribuição dos Conselhos Tutelares. O Congresso Nacional decreta Art. 1 Esta lei tem como objetivo alterar a Lei n 8. 069, de 13 de julho de 1990, dispondo sobre quantidade e critérios de distribuição dos Conselhos Tutelares. Art. 2º. O art. 132 da Lei nº 8. 069, de 13 de julho de 1990, passa a viger com a seguinte redação Art. 132. Em cada Município haverá, no mínimo, um Conselho Tutelar composto de cinco membros, escolhidos pela comunidade local para mandato de três anos, permitida uma recondução. Redação dada pela Lei nº 8. 242, de 12. 10. 1991 1º Para assegurar a eqüidade de acesso, caberá aos Municípios com mais de cem mil habitantes e ao Distrito Federal criar e manter Conselhos Tutelares, observada a proporção mínima de um Conselho para cada cem mil habitantes ou fração. 2º Quando houver mais de um Conselho Tutelar em um Município, caberá a este distribuí los conforme a configuração geográfica e administrativa da localidade, a população de crianças e adolescentes e a incidência de violações a seus direitos, assim como os indicadores sociais. 3º Cabe à legislação local a definição da área de atuação de cada Conselho Tutelar, devendo ser, preferencialmente, criado um Conselho Tutelar para cada região, circunscrição administrativa ou microrregião, observados os parâmetros indicados no 1º e no 2º. Art. 3º. Esta Lei entra em vigor da data de sua publicação. Sala das Sessões, em 04 de maio de 2011 ALESSANDRO MOLON Deputado Federal PT RJ JUSTIFICAÇÃO O presente projeto tem por objetivo aperfeiçoar o sistema de garantia de direitos brasileiro no que tange aos Conselhos Tutelares, órgãos essenciais ao referido sistema e vinculados aos municípios e ao Distrito Federal. O Conselho Nacional de Defesa dos Direitos da Criança e do Adolescente CONANDA reunido em assembléia em 17 de março de 2010, baixou a resolução nº 139 2010, dispondo exclusivamente sobre os Conselhos Tutelares. Tal medida deveu se à necessidade de atualização de outra resolução anterior, a de nº 75 2001, que estabeleceu os primeiros parâmetros de criação e funcionamento destes Conselhos. É fácil perceber que no interregno de uma resolução e outra, 10 anos se passaram, o que ocasionou natural, porém profunda discrepância principalmente no que tange ao número de Conselhos Tutelares necessários ao atendimento das necessidades da infância e da adolescência brasileiras e o previsto na legislação especial, em particular na Lei nº 8. 069, de 13 de julho de 1990, conhecida como ECA Estatuto da Criança e do Adolescente, mormente nas metrópoles e grandes cidades do país. A disposição atual da supracitada Lei estabelece que deve haver em cada município até 200 mil habitantes, no mínimo, um Conselho Tutelar, composto de cinco membros selecionados pela própria comunidade local, com mandato de três anos, e desde que atendidos os requisitos legais. Cumpre salientar que a nova Resolução do CONANDA tomou por base pesquisa que revelou a inexistência de Conselhos Tutelares em cerca de 10 dos Municípios brasileiros e graves deficiências no funcionamento da maioria dos já constituídos, informações aliadas aos Dados do IBGE que mostram que a população brasileira nos últimos 10 anos cresceu de 169. 590. 693 para 190. 755. 799, o que comprova a necessidade da atualização ora proposta neste projeto. Além de tratar se de atividade de interesse público relevante, as funções desenvolvidas pelos Conselhos Tutelares vão ao encontro das metas estabelecidas pelo Governo Federal na esfera dos Direitos Humanos, de forma a atender, proteger e defender integralmente os direitos da infância brasileira, principalmente considerando que cresce o número das que são maltratadas no país. Pelo exposto, cabe a esta Casa legislativa, dentro de sua esfera de atribuições, dar sua contribuição visando o bem estar de nossas crianças e adolescentes, adequando tão importante diploma legal à realidade nacional atual. Sala das Sessões, em 04 de maio de 2011. ALESSANDRO MOLON Deputado Federal PT RJ - PROJETO DE LEI Nº DE 2015 Do Sr. Aliel Machado Torna obrigatório o emprego do Regime Diferenciado de Contratações Públicas RDC previsto na Lei nº 12. 462, de 4 de agosto de 2011, para realização de licitações públicas, e dá outras providências. O Congresso Nacional decreta Art. 1º As licitações realizadas por órgãos e entidades integrantes da Administração Pública passam a ser efetuadas exclusivamente nos termos do Regime Diferenciado de Contratações Públicas RDC, previsto na Lei nº 12. 462, de 4 de agosto de 2001. Art. 2º Somente serão aplicados aos procedimentos referidos no art. 1º desta Lei os dispositivos da Lei nº 8. 666, de 21 de junho de 1993, cujo teor seja compatível com a Lei nº 12. 462, de 2001. Art. 3º O conjunto de normas destinado a reger as licitações e contratos administrativos, previsto no inciso XXVII do art. 22 da Constituição, passa a resultar da aplicação do disposto nos arts. 1º e 2º desta Lei. Art. 4º A Lei nº 12. 462, de 2001, passa a vigorar com as seguintes alterações Art. 1º É instituído o Regime Diferenciado de Contratações Públicas RDC aplicável a licitações realizadas por órgãos e entidades da Administração 2 Pública, qualquer que seja o objeto envolvido nesses procedimentos. NR Art. 13. As licitações serão realizadas sob a forma eletrônica. Parágrafo único. Os atos imputados aos licitantes serão praticados exclusivamente por meio eletrônico. NR Art. 4º Esta Lei entra em vigor na data de sua publicação e não se aplica a licitações cujos instrumentos convocatórios já tenham sido objeto de divulgação. Art. 5º Ficam revogados o 2º do art. 1º da Lei nº 12. 462, de 2011, a Lei nº 10. 520, de 17 de julho de 2002, e o art. 67 da Lei nº 9. 478, de 6 de agosto de 1997. JUSTIFICAÇÃO Apesar do inegável sucesso colhido na aplicação da Lei nº 10. 520, de 17 de julho de 2002, que viria a inspirar a edição do chamado Regime Diferenciado de Contratações Públicas conhecido pela sigla RDC continuam em pleno vigor e corriqueira utilização regras absolutamente ultrapassadas no que diz respeito à realização de licitações por órgãos e entidades integrantes da Administração Pública. Envelopes que há muito deixaram de ser lacrados, pelo menos no que diz respeito aos conluios que viabilizam, seguem servindo como instrumento para a esmagadora maioria dos contratos celebrados pelo Poder Público. O resultado desse contexto é aquele que se encontra em curso no âmbito da justiça federal paranaense. De fato, o caso da Petrobras trouxe ainda mais a lume a defasagem da Lei nº 8. 666, de 1993, na medida em que se alegou justamente o caráter obsoleto desse diploma para os atos ilícitos ali revelados. Justificaram se desvios de conduta os mais diversos com base na inadequação do Estatuto das licitações como meio de disciplinar as licitações da maior estatal brasileira. Por outro lado, as resistências que se verificaram no meio empresarial quanto à utilização do aludido RDC dão bem o diagnóstico da 3 relevância desse sistema. Foram inúmeras, em diversos veículos de comunicação, as veementes manifestações de juristas e especialistas a serviço de empreiteiras empenhadas em ferir a credibilidade do instrumento. Passados quase quatro anos desde a aprovação do mecanismo, contudo, essas vozes, sintomaticamente, não são ouvidas mais sequer em surdina. Nessa conjuntura, o projeto cuja aprovação se postula constitui justamente uma tentativa de ver disseminados e consolidados os eficazes procedimentos administrativos decorrentes do RDC, razão pela qual se pede a aprovação integral de seus termos. Sala das Sessões, em de de 2015. Deputado ALIEL MACHADO 2015_3748 --- # SentenceTransformer based on neuralmind/bert-large-portuguese-cased This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [neuralmind/bert-large-portuguese-cased](https://huggingface.co/neuralmind/bert-large-portuguese-cased) <!-- at revision aa302f6ea73b759f7df9cad58bd272127b67ec28 --> - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 1024 tokens - **Similarity Function:** Cosine Similarity <!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown --> ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("josedossantos/tree-txtInteiroTeorLimpo-bertimbau") # Run inference sentences = [ 'PROJETO DE LEI Nº DE 2015 Do Sr. Veneziano Vital do Rêgo Revoga dispositivos da Lei nº 12. 462, de 4 de agosto de 2011, para limitar a abrangência do Regime Diferenciado de Contratações aos casos inicialmente previstos pela lei original, e dá outras providências. O Congresso Nacional decreta Art. 1º Revogam se os incisos IV, V e VI, e o 3º do artigo 1º da Lei nº 12. 462, de 4 de agosto de 2011. Art. 2º Revogam se os artigos 1º e 2º da Lei nº 12. 873, de 24 de outubro de 2013. Art. 3º Revoga se o art. 15 A da Lei nº 12. 340, de 1º de dezembro de 2010. Art. 4º Esta Lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO A Lei nº 12. 462, de 4 de agosto de 2011, resultante da conversão em lei da Medida Provisória nº 527, de 2011, instituiu o Regime Diferenciado de Contratações Públicas RDC, regime criado para atender às obras para a Copa do Mundo e da Olimpíada. Ou seja, a diferenciação e as 2 facilidades previstas na referida lei foram previstas, excepcionalmente, para viabilizar as obras e serviços necessários para a realização dos grandes eventos em nosso País. Ressalta se que os objetivos iniciais acabaram sendo maculados, uma vez que foram inseridos no corpo da lei, e em outras leis, autorizações para a utilização permanente do RDC para finalidades bem distintas das originalmente previstas. Portanto, faltando pouco mais de um ano para a realização das Olimpíadas no Estado do Rio de Janeiro, entendemos que ao término do evento a Lei terá cumprido com os seus propósitos e não subsistirá razão para que qualquer outra obra venha a ser executada utilizando se do RDC como instrumento de licitação, voltando se a adotar integralmente, em toda a Administração Pública, os parâmetros da Lei nº 8. 666, de 21 de junho de 1993. Isto posto, solicitamos o apoio dos nobres colegas nesta Casa para aprovação do presente projeto de lei. Sala das Sessões, em de de 2015. Deputado VENEZIANO VITAL DO REGO 2015 9374', 'PROJETO DE LEI Nº DE 2015 Do Sr. Aliel Machado Torna obrigatório o emprego do Regime Diferenciado de Contratações Públicas RDC previsto na Lei nº 12. 462, de 4 de agosto de 2011, para realização de licitações públicas, e dá outras providências. O Congresso Nacional decreta Art. 1º As licitações realizadas por órgãos e entidades integrantes da Administração Pública passam a ser efetuadas exclusivamente nos termos do Regime Diferenciado de Contratações Públicas RDC, previsto na Lei nº 12. 462, de 4 de agosto de 2001. Art. 2º Somente serão aplicados aos procedimentos referidos no art. 1º desta Lei os dispositivos da Lei nº 8. 666, de 21 de junho de 1993, cujo teor seja compatível com a Lei nº 12. 462, de 2001. Art. 3º O conjunto de normas destinado a reger as licitações e contratos administrativos, previsto no inciso XXVII do art. 22 da Constituição, passa a resultar da aplicação do disposto nos arts. 1º e 2º desta Lei. Art. 4º A Lei nº 12. 462, de 2001, passa a vigorar com as seguintes alterações Art. 1º É instituído o Regime Diferenciado de Contratações Públicas RDC aplicável a licitações realizadas por órgãos e entidades da Administração 2 Pública, qualquer que seja o objeto envolvido nesses procedimentos. NR Art. 13. As licitações serão realizadas sob a forma eletrônica. Parágrafo único. Os atos imputados aos licitantes serão praticados exclusivamente por meio eletrônico. NR Art. 4º Esta Lei entra em vigor na data de sua publicação e não se aplica a licitações cujos instrumentos convocatórios já tenham sido objeto de divulgação. Art. 5º Ficam revogados o 2º do art. 1º da Lei nº 12. 462, de 2011, a Lei nº 10. 520, de 17 de julho de 2002, e o art. 67 da Lei nº 9. 478, de 6 de agosto de 1997. JUSTIFICAÇÃO Apesar do inegável sucesso colhido na aplicação da Lei nº 10. 520, de 17 de julho de 2002, que viria a inspirar a edição do chamado Regime Diferenciado de Contratações Públicas conhecido pela sigla RDC continuam em pleno vigor e corriqueira utilização regras absolutamente ultrapassadas no que diz respeito à realização de licitações por órgãos e entidades integrantes da Administração Pública. Envelopes que há muito deixaram de ser lacrados, pelo menos no que diz respeito aos conluios que viabilizam, seguem servindo como instrumento para a esmagadora maioria dos contratos celebrados pelo Poder Público. O resultado desse contexto é aquele que se encontra em curso no âmbito da justiça federal paranaense. De fato, o caso da Petrobras trouxe ainda mais a lume a defasagem da Lei nº 8. 666, de 1993, na medida em que se alegou justamente o caráter obsoleto desse diploma para os atos ilícitos ali revelados. Justificaram se desvios de conduta os mais diversos com base na inadequação do Estatuto das licitações como meio de disciplinar as licitações da maior estatal brasileira. Por outro lado, as resistências que se verificaram no meio empresarial quanto à utilização do aludido RDC dão bem o diagnóstico da 3 relevância desse sistema. Foram inúmeras, em diversos veículos de comunicação, as veementes manifestações de juristas e especialistas a serviço de empreiteiras empenhadas em ferir a credibilidade do instrumento. Passados quase quatro anos desde a aprovação do mecanismo, contudo, essas vozes, sintomaticamente, não são ouvidas mais sequer em surdina. Nessa conjuntura, o projeto cuja aprovação se postula constitui justamente uma tentativa de ver disseminados e consolidados os eficazes procedimentos administrativos decorrentes do RDC, razão pela qual se pede a aprovação integral de seus termos. Sala das Sessões, em de de 2015. Deputado ALIEL MACHADO 2015_3748', 'PROJETO DE LEI Nº DE 2011 Do Sr. Laércio Oliveira Altera a Lei nº 11. 901, de 12 de janeiro de 2009. O Congresso Nacional decreta Art. 1º A presente lei é editada visando incluir na norma competente a obrigatoriedade de contratação de bombeiro civil em edificações ou conjunto de edificações com fins comerciais. Art. 2º A Lei nº 11. 901, de 12 de janeiro de 2009, passa a vigorar acrescida do seguinte artigo Art. 10 A Toda e qualquer edificação ou conjunto de edificações públicas e particulares, comerciais ou assemelhadas, tem a obrigação de contratar bombeiro civil, devidamente qualificado, para fazer parte do quadro permanente de pessoal. Parágrafo único. Tal contratação poderá ser realizada, conforme o art. 2º da presente lei, por intermédio de empresa especializada em prestação de serviços de prevenção e combate a incêndio. NR Art. 3º Ficam revogadas todas as disposições que contrariam o que fora tratado na presente norma. Art. 4º Esta lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO Com a publicação da presente lei, acidentes como o ocorrido no município do Rio de Janeiro serão prevenidos. Isso porque, com a atuação permanente de um bombeiro civil, situações de perigo podem ser antecipadas e ações de evacuação de edificações comerciais em iminente risco de incêndio ou explosão ocorrerão de forma correta e prudente. O caso concreto citado foi provocado por desrespeito às normas de segurança e, também, por incompetência da entidade publica que emitiu alvará de funcionamento sem atentar às especificidades do edifício. Mas, mesmo assim, não podemos ignorar o fato de que se houvesse um profissional devidamente habilitado trabalhando permanentemente no quadro de pessoal daquele prédio as vítimas poderiam ter sido salvas. Este indivíduo teria a discernimento técnico suficiente para avaliar a situação e promover a imediata evacuação das pessoas ali presentes, procedendo às devidas cautelas contra o pânico. A fatalidade aconteceu porque nenhum dos cidadãos foi capaz de determinar que o vazamento de gás colocasse suas vidas em extremo risco. Sendo assim, com a publicação da presente lei, situações como essas poderão ser extintas. Portanto, nesse contexto, rogo aos meus pares apoio na aprovação integral da presente proposição. Sala das Sessões, em de de 2011. LAÉRCIO OLIVEIRA Deputado Federal PR SE', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 1024] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` <!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* --> ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 4,327 training samples * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code> * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | label | |:--------|:-------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 86 tokens</li><li>mean: 495.18 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 147 tokens</li><li>mean: 488.11 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>0: ~41.10%</li><li>1: ~58.90%</li></ul> | * Samples: | sentence_0 | sentence_1 | label | |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------| | <code>PROJETO DE LEI N. º DE 2016 Do Sr. Dr. Jorge Silva Institui a Semana nacional de atenção à saúde do homem. O Congresso Nacional decreta Art. 1º Fica instituída a Semana nacional de atenção à saúde do homem, a ser comemorada anualmente durante o mês de novembro. 1º A programação da Semana incluirá atividades educativas que visem à prevenção de doenças, à promoção da saúde e à conscientização sanitária, sendo abordados os agravos mais frequentes na população masculina de cada localidade, como doenças cardiovasculares, hipertensão arterial, diabetes, doenças do aparelho geniturinário e da próstata, neoplasias, andropausa, impotência, infertilidade, direitos sexuais e reprodutivos, doenças sexualmente transmissíveis, saúde mental, além de outras doenças e situações que afetem a saúde e o bem estar da população masculina. 2º Durante a Semana, as unidades de saúde públicas e privadas poderão oferecer à população masculina demonstrações, consultas, procedimentos diagnósticos, tratamentos e outras ações voltadas à promoção da saúde do homem. 4º As escolas de ensino médio poderão participar da Semana oferecendo atividades educativas sobre os agravos à saúde do homem mais comuns na região e promovendo debates sobre as questões sociais e culturais que distanciam a população masculina da esfera da saúde e do cuidado. Art. 2º Esta lei entra em vigor na data de sua publicação. 2 JUSTIFICAÇÃO O presente projeto de lei foi apresentado anteriormente pelo Deputado Jair Bolsonaro, em 2007, sendo aprovado nesta Casa em 2010, quando foi remetido para análise do Senado Federal. Na Casa Alta, foi aprovado na Comissão de Educação, Cultura e Esporte, tendo recebido a valiosa contribuição de emendas apresentadas pelo relator, Senador Armando Monteiro. A propositura, todavia, foi arquivada em dezembro de 2014, por estar em tramitação por duas legislaturas. Seu arquivamento, no entanto, não reflete a importância do tema ora abordado. De fato, a saúde do homem torna se questão cada vez mais premente no âmbito da saúde pública. Vários são os pontos relevantes acerca do tema, e que devem ser reiteradamente abordados como ação de educação em saúde. Além do aumento da incidência e da prevalência de doenças especificas, barreiras culturais dificultam o acesso dessa população aos serviços de saúde. Esse argumento foi muito bem abordado pelo autor original, de cuja justificação reproduzo alguns trechos bastante elucidativos Os homens são mais resistentes a buscar ajuda, não apenas nos problemas de saúde. São muito menos aplicados do que as mulheres no conhecimento e controle dos fatores de risco das doenças que os acometem com maior frequência. A ideia de instituir uma semana nacional da saúde masculina tem justamente o objetivo de ajudar os homens a superar estas dificuldades em cuidar da sua saúde. A comemoração de uma semana oficial dedicada aos principais problemas de saúde masculinos, nos âmbitos municipal, estadual e federal, sem dúvida contribuirá para despertar a sociedade e masculina para buscar melhor qualidade de vida. 3 Conhecendo melhor as doenças mais comuns, os fatores de risco, as medidas preventivas e, mesmo, onde encontrar ajuda, haverá maiores chances de que os índices de certas doenças diminuam ou, pelo menos, não aumentem inexoravelmente devido ao desconhecimento e à passividade. Precisamos mudar esta realidade e ampliar o acesso ao conhecimento, à prevenção e ao tratamento de doenças tipicamente masculinas. A semana proposta contribuirá para o aumento da consciência sobre os problemas masculinos, além de chamar a atenção das autoridades sanitárias para a necessidade de melhor organizar as ações e serviços de saúde pública nesta área. Por concordar com a posição exarada por seu nobre Autor, reapresento a propositura, acolhendo o teor das emendas então apresentadas no Senado Federal, e conto com o apoio de todos para sua apresentação. Sala das Sessões, em de de 2016. Deputado DR. JORGE SILVA 2016 10855. docx</code> | <code>1 4 8 1 B 2 F 1 9 4 9 4 8 1 B 2 F 1 9 4 9 PROJETO DE LEI Nº DE 2013 Do Sr. Dr. Jorge Silva Institui o mês Novembro Azul dedicado a ações de prevenção ao câncer de próstata e de promoção da saúde do homem. O Congresso Nacional decreta Art. 1º Fica instituído em todo o território Nacional o mês Novembro Azul dedicado à realização de ações de prevenção ao câncer de próstata e de promoção da saúde do homem. Art. 2º Nos meses de novembro de cada ano a critério dos gestores, em cooperação com a iniciativa privada, com entidades civis e organizações profissionais e científicas, realizará campanhas de esclarecimentos, exames e outras ações educativas e preventivas visando ao esclarecimento e incentivo à realização de exames preventivos para a detecção do câncer de próstata, assim como para outras doenças que acometem primordialmente a população masculina. Art. 3º Dentre as ações previstas, o Governo Federal deverá proceder à iluminação de locais públicos na cor azul. Art. 4º Esta Lei entra em vigor na data de sua publicação. 2 4 8 1 B 2 F 1 9 4 9 4 8 1 B 2 F 1 9 4 9 JUSTIFICAÇÃO O mês de outubro está sendo marcado por auspiciosa campanha, em todo o Brasil, de incentivo à prevenção do câncer de mama, conhecida como Outubro Rosa Essa campanha segue um movimento internacional e ganhou grande repercussão nos meios de comunicação, propiciando que mais e mais mulheres se sensibilizassem para a importância de se submeterem a exames e autocuidados voltados à prevenção da citada moléstia. Nesse mesmo sentido, o mês de novembro é internacionalmente dedicado às ações relacionadas ao câncer de próstata e à saúde do homem sendo que o dia 17 11 é o Dia Mundial de Combate ao Câncer de Próstata. Essa doença é o sexto tipo mais comum no mundo e o de maior incidência nos homens. As taxas da manifestação da doença são cerca de seis vezes maiores nos países desenvolvidos, com cerca de três quartos dos casos no mundo ocorrendo em homens com mais de 65 anos. Quando diagnosticado e tratado no início, tem os riscos de mortalidade reduzidos. No Brasil, é a quarta causa de morte por câncer e corresponde a 6 do total de óbitos para esse grupo. A próstata é uma glândula que se situa logo abaixo da bexiga e à frente do reto e tem como função produzir aproximadamente 70 do sêmen, representando um papel fundamental na fertilidade masculina. Dieta saudável, com menos gordura de origem animal, assim como atividade física, controle do peso, e diminuição do consumo de álcool e o não tabagismo ajudam a diminuir o risco do câncer. A partir dos 50 anos todos os homens devem procurar um serviço de saúde para realizar exames de rotina. Os sintomas mais comuns do tumor são a dificuldade de urinar, frequência urinária alterada ou diminuição da força do jato da urina, dentre outros. Quem tem histórico familiar da doença deve avisar o médico, que indicará os exames necessários. 3 4 8 1 B 2 F 1 9 4 9 4 8 1 B 2 F 1 9 4 9 Fica claro, portanto, que, a exemplo do que tem sido feito com tanto sucesso para a prevenção do câncer de mama nas mulheres, é importantíssimo que sejam realizadas atividades com vistas à promoção da saúde masculina, com foco na prevenção ao câncer de próstata. Isto posto, esperamos contar com o apoiamento de nossos ilustres Pares em ambas as Casas do Congresso Nacional para aprovar essa proposição que, com toda certeza, em muito contribuirá para a proteção da saúde da população masculina em nosso País. Sala das Sessões, em de de 2013. Deputado DR. JORGE SILVA 2013_25253</code> | <code>1</code> | | <code>CÂMARA DOS DEPUTADOS GABINETE DO DEPUTADO EDUARDO BOLSONARO PROJETO DE LEI No DE 2016. Do Sr. Eduardo Bolsonaro Altera dispositivos do Decreto Lei nº 3. 689, de 3 de outubro de 1941 Código de Processo Penal, para tornar obrigatória a conversão da prisão em flagrante em preventiva, nos casos que especifica, extinguindo as audiências de custódia. O Congresso Nacional decreta Art. 1º Os arts. 310 e 312 do Decreto Lei nº 3. 689, de 3 de outubro de 1941, passam a vigorar com a seguinte redação Art. 310 II converter a prisão em flagrante em preventiva, quando presentes os requisitos constantes do art. 312 deste Código, ou se revelarem inadequadas ou insuficientes as medidas cautelares diversas da prisão ou 1º Se o juiz verificar, pelo auto de prisão em flagrante, que o agente praticou o fato nas condições constantes dos incisos I a III do caput do art. 23 do Decreto Lei no 2. 848, de 7 de dezembro de 1940 Código Penal, poderá, fundamentadamente, conceder ao acusado liberdade provisória, mediante termo de comparecimento a todos os atos processuais, sob pena de revogação. 2º Para as providências de que trata este artigo, o juiz fundamentará sua decisão com base nas informações constantes no auto de prisão em flagrante, colhidas pela autoridade competente responsável pela sua lavratura, sendo que a apresentação do preso à autoridade judicial se dará, em sendo o caso, na respectiva audiência de instrução e julgamento. NR Art. 312 A prisão preventiva será decretada como garantia da ordem pública, da ordem econômica, por conveniência da instrução criminal, ou para assegurar a aplicação da lei penal, quando houver prova da existência do crime e indício suficiente de autoria. 1º A prisão preventiva também será decretada em caso de descumprimento de qualquer das obrigações impostas por força de outras medidas cautelares art. 282, 4º 2º Será obrigatória a conversão de prisão em flagrante em preventiva, desde que constatada qualquer das situações previstas nos incisos I a III do caput do art. 313 deste Código. NR Art. 2º Esta lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO A prática reiterada de atos criminosos gera sensação de impunidade que estimula os criminosos, apavora os cidadãos e acarreta aos policiais um sentimento de impotência, frente ao retrabalho diário a que estão submetidos esses profissionais. As audiências de custódia, instituídas pela Resolução nº 213, de 15 de dezembro de 2015, do Conselho Nacional de Justiça CNJ órgão integrante do Poder Judiciário, agravaram tal sensação ao estabelecer uma inversão de valores e papéis, pois os investigados passaram a ser, prioritariamente, os agentes policiais responsáveis pelas prisões, e os criminosos de fato foram travestidos de vítimas em potencial, independente da natureza ou gravidade da infração penal praticada. À inconstitucionalidade do ato normativo formalizado pelo CNJ, ao usurpar a competência legislativa do Congresso Nacional, inovando em institutos e procedimentos processuais não previstos em nosso ordenamento jurídico devem ser ministrados os remédios legislativos e judiciais cabíveis, considerando as devidas proposições e searas de competência. Entretanto, cumpre nos observar que a operacionalização de tais procedimentos se pauta, mesmo que inconstitucionalmente, nas subjetividades existentes em nosso Código de Processo Penal, que submetem à apreciação do juiz as situações em que será cabível a decretação da prisão preventiva, em detrimento da aplicação de medidas cautelares diversas da prisão ou mesmo da concessão de liberdade provisória ao indivíduo preso em flagrante delito. Diante do exposto, o presente projeto busca estabelecer, de forma objetiva, as situações em que será obrigatória a decretação da prisão preventiva, principalmente quando da conversão de prisão em flagrante naquela modalidade, objeto das fatídicas audiências de custódia recentemente adotadas no Brasil. A alteração proposta no inciso II do art. 310 do Código de Processo Penal CPP reside na substituição da conjunção aditiva e pela conjunção alternativa ou Com a redação atual, para a decretação da prisão preventiva, mesmo estando presentes os requisitos constantes do art. 312 do CPP, há que se observar, a critério subjetivo do julgador, se seriam adequadas ou suficientes as medidas cautelares diversas da prisão. Ou seja, mesmo presentes os requisitos para a manutenção da restrição de liberdade, pode se aplicar tão somente uma medida cautelar, ficando a sociedade novamente à mercê de criminosos. A redação ora apresentada, ao estabelecer a alternância, impõe a decretação da prisão preventiva quando cumpridos os requisitos e possibilita sua utilização quando não se mostrarem adequadas ou suficientes as medidas cautelares diversas da prisão, nos casos em que são inicialmente aplicadas. O acréscimo do 2º ao artigo 310 busca preservar a autenticidade e legalidade dos atos praticados pelos integrantes dos organismos policiais quando da realização de sua atividade laboral, os quais foram totalmente desprezados pela Resolução nº 213 2015 CNJ. Da forma proposta na operacionalização das audiências de custódia, independente do que se aferiu em investigações policiais e lavraturas de autos de prisão em flagrante, o simples testemunho do preso pode alterar a dinâmica dos fatos, transformando profissionais de segurança pública em torturadores criminosos. O profissionalismo, o comprometimento, a qualidade de representantes do Estado, a competência e legitimidade definidas em lei dos agentes de segurança pública não devem ser desprezadas por uma Resolução do CNJ que determina a realização de audiências de custódia Os atos praticados pelos policiais que realizam uma prisão em flagrante, que investigam determinada infração penal e, principalmente, pela autoridade policial que preside a lavratura do necessário auto, devem ser legitimados pelo poder público, sob pena de falência de todo o sistema de segurança pública, por se priorizar com exclusividade a análise por parte de um juiz singular em audiência de custódia. Dessa forma, tendo sido colhidas e documentadas as informações necessárias quando da lavratura do auto de prisão, o juiz deverá nelas se basear para a tomada de decisão, o que, inclusive, encontra se devidamente previsto na norma legal vigente. A alteração no caput do art. 312 busca retirar a atual possibilidade de decretação da prisão preventiva e vincular a autoridade judicial a adotar tal procedimento, estando presentes os requisitos necessários. Quanto à conversão de prisão em flagrante em preventiva, passa a ser obrigatória, desde que presente uma das situações objetivas previstas nos incisos do art. 313. Não podemos coadunar com as políticas de depreciação das instituições públicas patrocinadas em nosso país. Na realidade brasileira, enquanto policiais são tratados indistintamente como criminosos, as diretrizes estabelecidas pelo governo federal pregam o assistencialismo com fins eleitoreiros e a proteção a criminosos das mais variadas espécies, chegando se ao absurdo da apresentação, por parte de um parlamentar do PT, de Projeto de Lei criando o Estatuto do Presidiário garantindo regalias absurdas para criminosos condenados como creme hidratante, xampu, visita íntima, salão de beleza, além de cinco médicos, três enfermeiros, três odontólogos, três psicólogos, três nutricionistas, seis técnicos em higiene dental, seis auxiliares de enfermagem, dentre outros, a cada grupo de 400 presos. Conto com os nobres pares para a aprovação do presente Projeto de Lei, que visa a reduzir efetivamente a sensação de impunidade que vigora no Brasil. Sala das Sessões, em de fevereiro de 2016. EDUARDO BOLSONARO Deputado Federal PSC SP</code> | <code>PROJETO DE LEI Nº DE 2015. Do Sr. Laerte Bessa Altera o Código de Processo Penal para estabelecer a audiência de custódia, e dá outras providências. O CONGRESSO NACIONAL decreta Art. 1º Os arts. 304, 306, 310 e 322 do Decreto Lei nº 3. 689, de 3 de outubro de 1941 Código de Processo Penal, passam a vigorar com as seguintes alterações Art. 304................................................................................................ 1o Resultando das respostas fundada a suspeita contra o conduzido, a autoridade mandará recolhê lo à prisão, exceto no caso de livrar se solto, de aplicação de medida cautelar diversa da prisão ou de manifesta causa excludente de ilicitude prevista no art. 23 do Decreto Lei no 2. 848, de 7 de dezembro de 1940 Código Penal, e prosseguirá nos atos do inquérito ou processo, se para isso for competente se não o for, enviará os autos à autoridade que o seja. NR 4º O preso tem o direito de ser assistido por advogado, público ou contratado, durante o seu interrogatório policial, podendo ser nomeado defensor dativo pelo delegado de polícia, com a concordância do interrogando, se este não o dispensar expressamente ou permanecer em silêncio, no exercício da autodefesa. NR 5º Todo preso será submetido a exame de corpo de delito cautelar, realizado por perito médico oficial, onde houver, ou médico nomeado pelo delegado de polícia, preferencialmente entre os integrantes da rede pública de saúde. NR Art. 306.................................................................................................. 3º A comunicação da prisão será acompanhada do laudo preliminar do exame de corpo de delito cautelar do preso. NR 4º O preso não poderá permanecer em delegacia ou unidade da polícia judiciária por prazo superior a 72 horas, devendo ser transferido para setor específico de custódia, triagem e transição de presos ou para estabelecimento penitenciário, à disposição do juiz. NR Art. 310.................................................................................................. 2º No prazo de 24 horas, após o recebimento do auto de prisão em flagrante, o juiz poderá designar audiência de custódia, que será realizada com a participação da defesa e do Ministério Público, caso o indiciado tenha sido interrogado sem defensor ou não tenha sido posto liberdade pelo delegado de polícia, mediante fiança ou outra medida cautelar diversa da prisão. NR 3º A audiência de custódia terá por objetivo as providências elencadas no caput e poderá ser realizada por videoconferência. NR Art. 322 No caso de prisão em flagrante, o delegado de polícia poderá colocar o indiciado em liberdade aplicando, em decisão fundamentada, isolada ou cumulativamente, as medidas cautelares diversas da prisão previstas nos incisos I, II, III, IV, V, VIII e IX do art. 319, nas seguintes hipóteses I crimes culposos II crimes dolosos punidos com pena de detenção III crimes dolosos punidos com pena de reclusão não superior a quatro anos IV quanto não estiverem presentes os requisitos da prisão preventiva. NR Parágrafo único. O delegado de polícia comunicará o juiz competente nos termos do art. 306, que decidirá, na forma do art. 310, sobre o auto de prisão em flagrante e as medidas cautelares eventualmente aplicadas. NR Art. 2º Esta Lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO Muito se tem falado sobre audiência de custódia, alguns antevendo na medida a solução para o elevado número de presos provisórios e o fim de todas as formas de violações aos direitos dos presos. Por essa razão, importante notar que não existe uma suposta imprescindibilidade de apresentação de todos os presos em flagrante exclusivamente ao juiz de direito. O Brasil é signatário do Pacto Internacional Sobre Direitos Civis e Políticos, da Organização das Nações Unidas, cujo item 03 do art. 09 dispõe da seguinte forma, verbis ARTIGO 9 3. Qualquer pessoa presa ou encarcerada em virtude de infração penal deverá ser conduzida, sem demora, à presença do juiz ou de outra autoridade habilitada por lei a exercer funções judiciais e terá o direito de ser julgada em prazo razoável ou de ser posta em liberdade. A prisão preventiva de pessoas que aguardam julgamento não deverá constituir a regra geral, mas a soltura poderá estar condicionada a garantias que assegurem o comparecimento da pessoa em questão à audiência, a todos os atos do processo e, se necessário for, para a execução da sentença. Grifamos. A Convenção Americana sobre Direitos Humanos Pacto de São José da Costa Rica dispõe de forma semelhante no item 05 do art. 07, vejamos ARTIGO 7 5. Toda pessoa detida ou retira deve ser conduzida, sem demora, à presença de um juiz ou outra autoridade autorizada pela lei a exercer funções judiciais e tem direito a ser posta em liberdade, sem prejuízo de que prossiga o processo. Sua liberdade pode ser condicionada a garantias que assegurem o seu comparecimento em juízo. Grifamos. Esses dispositivos são corroborados pela Resolução nº 43 73, da Assembleia Geral, de 9 de dezembro de 1988, da Organização das Nações Unidas, traz o Conjunto de Princípios para a Proteção de Todas as Pessoas Sujeitas a Qualquer Forma de Detenção ou Prisão dentre os quais se destacam os seguintes PRINCÍPIO 4 As formas de detenção ou prisão e as medidas que afetem os direitos do homem, da pessoa sujeita a qualquer forma de detenção ou prisão devem ser decididas por uma autoridade judiciária ou outra autoridade, ou estar sujeitas a sua efetiva fiscalização. Grifamos. PRINCÍPIO 37 A pessoa detida pela prática de uma infração penal deve ser presente auma autoridade judiciária ou outra autoridade prevista por lei, prontamente após sua captura. Essa autoridade decidirá sem demora da legalidade e necessidade da detenção Grifamos. Da simples leitura dos dispositivos depreende se que o detido deve ser conduzido sem demora à presença de um juiz ou outra autoridade autorizada por lei a exercer funções judiciais. Para que não reste dúvida, a Resolução nº 43 73 da ONU traz o conceito da expressão outra autoridade como outra autoridade estabelecida nos termos da lei cujo estatuto e mandato ofereçam as mais sólidas garantias de competência, imparcialidade e independência Analisando esse conceito sob a ótica do ordenamento jurídico interno, percebe se que a lei maior da República instituiu o cargo de delegado de polícia como dirigente das Polícias Civis, sendo, portanto, o titular das funções de polícia judiciária e de apuração de infrações penais, nos termos do no 4º do art. 144, da Constituição Federal. Nesse sentido, quando restringe determinado ato à reserva de jurisdição, a Constituição Federal o faz claramente, como no inciso LXI do art. 5º, ao dispor que ninguém será preso senão em flagrante delito ou por ordem escrita e fundamentada de autoridade judiciária competente. A análise da legalidade da prisão em flagrante delito, portanto, não está sujeita à reserva de jurisdição. Corroborando esse entendimento, o inciso LXII do art. 5º da Carta Magna estabelece que a prisão de qualquer pessoa e o local onde se encontre serão comunicados imediatamente ao juiz competente e à família do preso ou à pessoa por ele indicada não impondo, pois a apresentação imediata do preso ao juiz, que deve ser comunicado imediatamente. Da mesma forma, o inciso LXIV do art. 5º da Constituição dispõe que o preso tem direito à identificação dos responsáveis por sua prisão ou por seu interrogatório policial Percebe se que a Constituição Federal serve ao mesmo tempo de autorização constitucional para que o delegado de polícia seja o primeiro a decidir sobre a prisão e garantia do preso para que tenha ciência sobre a identidade de quem o prendeu e da autoridade de polícia judiciária responsável pela análise da legalidade do ato que restringiu sua liberdade. Portanto, o delegado de polícia é a autoridade autorizada pela Constituição Federal e por diversas leis federais a exercer atipicamente funções judiciais, mesmo não integrando o Poder Judiciário, o fazendo, por exemplo, quando arbitra fiança como condição para concessão da liberdade do preso em flagrante, quando apreende um bem relacionado ao crime, quando homologa a prisão em flagrante e determina o recolhimento do conduzido à prisão ou quando promove o indiciamento, ato que se reveste das mesmas formalidades das decisões judiciais, nos termos do 6º, do art. 2º, da Lei Federal nº 12. 830, de 20 de junho de 2013, verbis O indiciamento, privativo do delegado de polícia, dar se á por ato fundamentado, mediante análise técnico jurídica do fato, que deverá indicar a autoria, materialidade e suas circunstâncias Podemos citar diversos outros dispositivos legais que habilitam o delegado de polícia a exercer tais funções. Basta notar que o ingresso no cargo de delegado de polícia federal possui requisitos semelhantes aos exigidos pela Constituição Federal1 para ingresso na magistratura, nos termos da Lei nº 9. 266 96 2, alterada pela Lei nº 13. 047 2014. A Lei Federal nº 12. 830 2013, por sua vez, estabelece regras que garantem a autonomia do delegado de polícia em sua atividade, tais como a exigência de despacho fundamentado com base no interesse público ou descumprimento de normas regulamentares para que o inquérito possa ser avocado art. 2º, 4º a necessidade de fundamentação para que o delegado possa ser removido de sua lotação art. 2º, 5º a previsão de que o indiciamento é ato privativo do delegado de polícia art. 2º, 6º além de lhe ser deferido o mesmo tratamento protocolar dispensado a magistrados, promotores e advogados art. 2º, 7º A interpretação lógica, sistemática e teleológica dos dispositivos analisados nos permite concluir que as funções exercidas pelo delegado de polícia encontram não só amparo, mas verdadeira previsão no Pacto Internacional Sobre Direitos Civis e Políticos e na Convenção Americana de Direitos Humanos. Por essas razões, é possível concluir que o delegado de polícia é, ao lado do juiz, a autoridade habilitada por lei a tomar conhecimento da prisão e decidir sobre sua legalidade. Quanto à análise da conveniência de se instituir a audiência de custódia em que o preso deve ser apresentado ao juiz e não ao delegado de polícia, existe a fundada dúvida da necessidade de se conduzir todos os presos indistintamente à presença do juiz. 1 CF. Art. 93.... I ingresso na carreira, cujo cargo inicial será o de juiz substituto, mediante concurso público de provas e títulos, com a participação da Ordem dos Advogados do Brasil em todas as fases, exigindo se do bacharel em direito, no mínimo, três anos de atividade jurídica e obedecendo se, nas nomeações, à ordem de classificação 2 Art. 2 o B. O ingresso no cargo de Delegado de Polícia Federal, realizado mediante concurso público de provas e títulos, com a participação da Ordem dos Advogados do Brasil, é privativo de bacharel em Direito e exige 3 três anos de atividade jurídica ou policial, comprovados no ato de posse. Certamente é mais racional, lógico e menos oneroso em vários sentidos ampliar as possibilidades de o delegado de polícia conceder fiança e até mesmo aplicar algumas medidas cautelares não sujeitas à reserva de jurisdição, logo após a prisão em flagrante, reservando a audiência de custódia para os casos de autuados por crimes mais graves que permanecerem presos. A par disso, existem sérias preocupações com relação ao sistema acusatório. Em função do princípio acusatório, é necessário um distanciamento seguro entre o julgador e os fatos incipientes da investigação, pois meras conjecturas iniciais não contrastadas com outros elementos oriundos da investigação de seguimento não raramente conduzem até a pessoa mais cautelosa a equívocos. Adotando se um modelo de audiência de custódia, é inevitável que o julgador cristalize um juízo acerca da conduta do preso com base exclusivamente naqueles elementos colhidos precariamente, sem contraponto ou cotejo com outros que confirmem ou não as conjecturas iniciais da prisão, gerando um convencimento embasado em indícios de autoria e materialidade muitas vezes frágeis, que inevitavelmente acompanharão o magistrado até a sentença. Esse juízo de valor sobre os fatos brutos por parte do juiz representa um retrocesso grave em termos de garantias individuais e mácula sobre os pressupostos de isenção e imparcialidade do julgador, valores inexoráveis do sistema acusatório, ao lado da separação das funções de investigar, acusar e julgar. Daí a importante da função do delegado de polícia, que se assemelha àquela exercida pelo juiz de instrução, visto que evita essa contaminação do juiz julgador. Nesse diapasão, quem conhece a sistemática procedimental da prisão em flagrante, sabe que o delegado de polícia não é mero homologador de prisões, porquanto faz a análise da situação sob os aspectos fáticos e jurídicos, não estando obrigado a homologar prisões ilegais ou que não estejam em situação flagrancial. Por outro lado, uma lei obrigando a apresentação de todos os presos perante o Poder Judiciário exigiria a organização de volumosas pautas de audiência com juiz, Ministério Público e Defensoria Pública diariamente, e não apenas nos dias úteis durante o horário de expediente, sob pena de marcante incoerência. Se pensarmos nos milhares de municípios existentes no Brasil, a proposta de audiência de custódia se mostra totalmente impraticável, visto que em sua grande maioria conta apenas com um delegado de polícia, sendo esta muitas vezes a única autoridade a menos de 200 km de distância que, por dever de ofício, vai até onde se encontra o cidadão detido para analisar a legalidade de sua prisão, postura que dificilmente se verá por parte de uma comissão de audiência de custódia, visto que hoje nem mesmo um defensor público é disponibilizado para acompanhar um simples auto de prisão em flagrante. Não é preciso grande esforço para perceber que uma lei com tais imposições não seria observada, criando mais uma causa de nulidade processual da prisão, que resultaria inevitavelmente no relaxamento de prisões em massa, trazendo intranquilidade social e depreciação à imagem do Poder Judiciário. Não se olvida, ainda, que a audiência de custódia aumentaria os gastos públicos na área de segurança pública, onde os recursos são notoriamente escassos, exigindo a contratação de grandes efetivos policiais ou, o que é mais provável, o desvio dos poucos policiais para atividades alheias às suas funções, como o transporte e a custódia de presos. Para os magistrados e auxiliares da Justiça não seriam insignificantes os riscos pessoais, porquanto se veriam em meio a inúmeros presos, fazendo de Fóruns verdadeiras cadeias transitórias, diante do tráfego de presos para realização de audiências de custódia, podendo ensejar tentativas de resgates violentos, notadamente em casos de integrantes de organizações criminosas, o que, por consequência, exigiria mais segurança, mais recursos e contratações, onerando ainda mais os contribuintes. Portanto, buscando conciliar os ditames constitucionais e dos tratados de direitos humanos ao sistema processual penal brasileiro, a fim de extirpar qualquer possibilidade de violações à incolumidade do preso ficarem impunes, apresentamos o presente projeto, que objetiva promover as seguintes medidas 1 Garantir a efetiva participação da defesa técnica no interrogatório policial e de que todos os presos sejam submetidos a exame de corpo de delito 2 A vedação à custódia de presos em delegacias por prazo superior a 72 horas, visto que tal situação, ainda persistente no Brasil, representa uma situação degradante violadora dos direitos humanos previstos em tratados internacionais, pois o preso não exerce direitos básicos previstos na Lei de Execução Penal 3 Possibilidade de o delegado de polícia aplicar algumas medidas cautelares não sujeitas à reserva de jurisdição, servindo como alternativa à manutenção de prisões desnecessárias por crimes leves 4 O estabelecimento da audiência de custódia no prazo de 24 horas, após o recebimento do auto de prisão em flagrante, para preso em flagrante que não tenha sido colocado em liberdade previamente, com fiança ou outra medida cautelar diversa da prisão. Firme no propósito de colaborar no aperfeiçoamento do ordenamento jurídico e apresentar soluções razoáveis e eficazes, ofertamos o presente projeto, que temos a certeza significará um avanço notório na proteção aos direitos humanos. Sala das sessões, 24de fevereiro de 2015. LAERTE BESSA Deputado Federal</code> | <code>1</code> | | <code>PROJETO DE LEI Nº DE 2011 Do Sr. Deputado GABRIEL CHALITA e outros Acrescenta o parágrafo 6º no art. 392A da Consolidação das Leis do Trabalho CLT, aprovada pelo Decreto Lei nº 5. 452, de 1º de maio de 1943 altera o caput do art. 71A da Lei nº 8. 213, de 24 de julho de 1991, que dispõe sobre os Planos de Benefícios da Previdência Social e dá outras providências. O Congresso Nacional decreta Art. 1º. O art. 392A da Consolidação das Leis do Trabalho, aprovada pelo Decreto Lei nº 5. 452, de 1º de maio de 1943, passa a vigorar acrescido do seguinte 6º Art. 392 A. 6º A concessão contida no caput do artigo fica estendida ao empregado que, sozinho, adotar ou obtiver guarda judicial para fins de adoção de criança, excluído o disposto no art. 473, III da Consolidação das Leis do Trabalho, art. 7º, XIX da Constituição Federal e art. 10, II, 1º do ADCT. NR Art. 2º. O art. 71A da Lei nº 8. 213, de 24 de julho de 1991, que dispõe sobre os Planos de Benefícios da Previdência Social e dá outras providências, passa a vigorar com a seguinte redação e acrescido do 2º, renumerando se o parágrafo único para 1º Art. 71A. À segurada da Previdência Social que adotar ou obtiver guarda judicial para fins de adoção de criança é devido salário maternidade pelo prazo de 120 cento e vinte dias 2º. O disposto no caput do artigo fica estendido ao segurado da Previdência Social que, sozinho, adotar ou obtiver guarda judicial para fins de adoção de criança. NR Art. 3º. Esta Lei entra em vigor na data de sua publicação. JUSTIFICATIVA A nova Lei de Adoção entrou em vigor em novembro de 2009 com um leque de mudanças sobre o tema adoção em geral, importando alterações desde os artigos 1618 e 1619 do novo Código Civil, Estatuto da Criança e do Adolescente, na Lei de Investigação de Paternidade como até na CLT Consolidação das Leis do Trabalho. Em relação aos direitos da criança, a nova Lei traz significativas novidades em diversos pontos, mas, que, em certos momentos ainda necessitam a continuidade dos debates com aprofundamento dos temas para que as soluções sejam encontradas diante dos obstáculos ainda instituídos dentro dos procedimentos utilizados na tramitação das demandas sócio familiares. Para tanto foi instituída a Frente Parlamentar Mista Intersetorial em Defesa das Políticas de Adoção e da Convivência Familiar e Comunitária, a fim de propor, discutir, aperfeiçoar, implementar, incentivar e acompanhar políticas públicas em defesa da adoção e que tem entre seus coordenadores, os proponentes do presente projeto de lei abaixo elencados. A proposta em tela objetiva, primeiramente, sanar a omissão no dispositivo legal à época da edição da Lei nº 12. 010, de 3 de agosto de 2009, que dispõe sobre adoção, altera as Leis nºs 8. 069, de 13 de julho de 1990 Estatuto da Criança e do Adolescente, 8. 560, de 29 de dezembro de 1992 revoga dispositivos da Lei nº 10. 406, de 10 de janeiro de 2002 Código Civil, e da Consolidação das Leis do Trabalho CLT, aprovada pelo Decreto Lei nº 5. 452, de 1º de maio de 1943, e dá outra providências, o qual deixou de revogar parcialmente ou alterar a redação do disposto sobre o salário maternidade. Assim, a propositura visa, especificamente, tratar de forma idêntica pessoas que adotam crianças e adolescentes, pois o atual artigo 71A da Lei nº 8. 213, de 24 de julho de 1991, como dito acima, ainda estabelece a diferença de período para a concessão do salário maternidade relacionada à idade do adotado, diferentemente do texto atual da CLT Consolidação das Leis do Trabalho, que já teve revogação expressa pela nova Lei de Adoção Lei nº 12. 010 09 eliminando referida discriminação. Demanda essa que se dá, atualmente, com a constatação de que a exclusão do sistema de direitos da chamada adoção necessária ou tardia, acaba por servir, ainda mais, a desestimular a prática ainda tão necessária de atenção e incentivo de políticas públicas. Com a presente proposição fica com o mesmo período de gozo da licença maternidade e respectiva remuneração por meio do salário maternidade toda trabalhadora que adotar criança ou adolescente de qualquer idade, vez que o Estatuto da Criança e do Adolescente, assim dispõe em seu artigo 2º caput Art. 2º. Considera se criança, para os efeitos desta Lei, a pessoa até doze anos de idade incompletos e adolescentes aquele entre doze e dezoito anos de idade É sabido por todos, notadamente pelo Poder Público, que as instituições de acolhimento institucional no país têm como perfil crianças maiores de um ano de idade, adolescentes e grupos de irmãos, sem contar os portadores de grave doença e as crianças e adolescentes com deficiência, todos estes justamente excluídos da legislação que ora se pretende alterar, vez que o estágio de convivência é crucial para a formação da nova família. Cabe notar, ainda, que segundo os últimos dados apresentados pelo Conselho Nacional de Justiça 31 08 2011 dos 27. 478 interessados em adotar inscritos no Cadastro Nacional de Adoção, 22. 451 deixaram claro o desejo de adotar crianças com até 1 ano de idade e 22. 702 manifestaram o desejo de por apenas uma criança. Assim, o sistema de direitos deve atender presente acionamento, a fim de impedir que referida diferenciação injusta e desvinculada das reais necessidades da demanda do novo núcleo familiar que se forma, se torne mais um obstáculo à decisão da prática da adoção. Importante notar, ainda, que a convenção sobre os Direitos da Criança, Parte I, Artigo I, a seguir in verbis, ratificada pelo Brasil por meio do Decreto nº 99. 710, de 21 de novembro de 1990, não diferencia criança de adolescente e deverá ser utilizada como paradigma para a presente proposição PARTE I ARTIGO I Para efeitos da presente Convenção considera se como criança todo ser humano com menos de dezoito anos de idade, a não ser que, em conformidade com a lei aplicável à criança, a maioridade seja alcançada antes. Dessa forma, a inclusão de isonomia do período de licença maternidade e a concessão do respectivo salário maternidade para as adotantes de crianças com idade superior a um ano e adolescentes visa garantir, igualmente, todos os estímulos essenciais ao estabelecimento do vínculo afetivo entre as partes envolvidas, possibilitando o exato cumprimento do disposto constitucional garantido pela Carta Magna em seu artigo 7º, inciso XVIII. Ademais, há que se considerar que quanto maior a idade e respectivamente maior o tempo de institucionalização da criança e do adolescente, maior o desafio de romper os traumas adquiridos na trajetória de cada uma das vidas que acabaram por ali estarem presentes. Outro ponto abordado na presente propositura relaciona se a concessão dos mesmos direitos da mãe adotante ao pai sozinho que adota, ressaltando, porém, a complementação do disposto fazendo menção à exclusão, no caso em tela, da licença paternidade já disposta na Consolidação das Leis do Trabalho, Constituição Federal e ADCT, a fim de se evitar o entendimento de somatória de direitos, apesar de não se confundirem por se tratarem de naturezas distintas. Cumpre nos esclarecer que o caso aqui não se refere a licença paternidade, pois esta se pressupõe à existência de uma licença maternidade, seja através do nascimento de um filho para pais casados ou não ou adotados. Vimos que essa licença maternidade foi estendida às mães adotantes, pois a Constituição Federal estabeleceu que não existe diferença entre filhos e, com isso, consequentemente, a licença paternidade também é estendida aos pais que adotam conjuntamente com a mãe. Quando uma mulher solteira adota, ela tem, naturalmente, o direito a licença maternidade também, mas um pai solteiro que adota sozinho não tem a mesma proteção legal. Ele só teria direito a licença paternidade de 5 cinco dias disposta no art. 473, III da Consolidação das Leis do Trabalho, art. 7º, XIX da Constituição Federal e art. 10, II, 1º do ADCT. Assim, verificamos que a natureza da licença paternidade de 5 cinco dias disposta na legislação está instrinsecamente ligada a existência da licença maternidade para adotados ou não porque a criança estaria sob a proteção do período correlacionado ao benefício materno, ou seja, de maior tempo de permanência, pois ninguém é pai naturalmente sem uma mãe. Mas quando a adoção é advinda apenas de forma paterna, sozinha, não há proteção legal para esse período de amparo a esta criança. Assim a jurisprudência vem dando a extensão desse direito ao homem que adota sozinho. Dessa forma o PL em tela visa eliminar essa omissão na legislação. Em relação ao campo de escolha para a extensão do direito ora tratado, explicitamos que, não poderia se misturar no âmbito da licença paternidade que apresenta outra natureza. Como se trata apenas de se estender direitos não vislumbramos problema em constá la neste campo para não adentrarmos na contra mão da Constituição Federal que igualou todos os filhos, estabelecendo que não existe diferença na maternidade advinda naturalmente ou pela adoção, tendo o legislador à época também não configurado respectivo direito em matéria separada. Nesse sentido, pela realidade social ainda se apresentar de maneira tão distante na busca da igualdade de oportunidades para a vida dessas crianças e adolescentes, pretende se com a presente propositura sanar a insegurança jurídica que a atual legislação apresenta pela já revogação expressa do artigo relacionado à licença maternidade, presente na CLT ao abolir a diferença estabelecida anteriormente no que se referia às idades dos adotados, mas sem ter sido modificada a legislação previdenciária, a fim de aperfeiçoar a disciplina da maternidade em caso de adoção de crianças em geral e adolescentes, como estender referida concessão aos empregados que, sozinhos, adotam ou obtém guarda judicial para fins de adoção, na esteira dos fundamentos apresentados e na melhor forma de direito. Em virtude do exposto, convidamos os nobres pares a aprovar com celeridade a proposição. Sala das Sessões, em de de 2011. GABRIEL CHALITA ALESSANDRO MOLON Deputado Federal PMDB SP Deputado Federal PT RJ REGUFFE Deputado Federal PDT DF</code> | <code>CÂMARA DOS DEPUTADOS PROJETO DE LEI Nº DE 2011. Da Sra. Andreia Zito Concede ao pai adotivo solteiro, o direito à licença paternidade e ao salário paternidade, alterando a consolidação das Leis do Trabalho, aprovada pelo Decreto Lei nº 5. 452, de 1º de maio de 1943, e a Lei nº 8. 213, de 24 de julho de 1991. O Congresso Nacional decreta Art. 1º A Consolidação das Leis do Trabalho, aprovada pelo Decreto Lei nº 5. 452, de 1º de maio de 1943, passa a vigorar acrescida do seguinte dispositivo Art. 392 B. Ao empregado que adotar ou obtiver guarda judicial para fins de adoção de criança, desde que seja solteiro, será concedida licença paternidade nos termos abaixo especificados 1º No caso de adoção ou guarda judicial de criança até 1 um ano de idade, o período de licença será de 120 cento e vinte dias. 2º No caso de adoção ou guarda judicial de criança a partir de 1 um ano até 4 quatro anos de idade, o período de licença será de 60 noventa dias. 3º No caso de adoção ou guarda judicial de criança a partir de 4 quatro anos até 8 oito anos de idade, o período de licença será de 30 trinta dias. 4º A licença paternidade só será concedida mediante apresentação do termo judicial de guarda ao adotante ou guardião. Art. 2º A Lei nº 8. 213, de 24 de julho de 1991, passa a vigorar acrescida do seguinte dispositivo Art. 71 B. Ao segurado da Previdência Social que adotar ou obtiver guarda judicial para fins de adoção de criança é devido salário paternidade pelo período de 120 cento e vinte dias, se a criança tiver até 1 um ano de idade, de 60 sessenta dias, se a criança tiver entre 1 um e 4 quatro Anos de idade, e de 30 trinta dias, se a criança tiver de 4 quatro a 8 oito anos de idade. CÂMARA DOS DEPUTADOS Art. 3º No caso dos segurados da previdência social adotantes, a alíquota para o custeio das despesas decorrentes desta Lei será a mesma que custeia as seguradas gestantes, disposta no inciso I do art. 22 da Lei nº 8. 212, de 24 de julho de 1991. Art. 4º Esta Lei entra em vigor na data de sua publicação. JUSTIFICAÇÃO O presente projeto de lei tem por objetivo aplicar de forma direta, um dos princípios basilares estabelecidos na Constituição Brasileira. De acordo com o Art. 5º da Constituição Federal, que assim diz Todos são iguais perante a lei, sem distinção de qualquer natureza, garantindo se aos brasileiros e aos estrangeiros residentes no País a inviolabilidade do direito à vida, à liberdade, à igualdade, à segurança e à propriedade, nos termos seguintes I homens e mulheres são iguais em direitos e obrigações, nos termos desta Constituição... são os motivos que nos levam a apresentar esta proposição com a justificação que entendo ser necessária para a sensibilização de todos os nobres parlamentares. Em pleno século XXI, podemos entender que os modos e pensamentos, como também as ações dos cidadãos, independentemente, se homem ou mulher, estão cada vez mais se tornando isonômicas, sob os mais diversos aspectos. Em 1988, data da promulgação da Carta Magna, ainda no século XX, no tocante aos direitos constitucionais dos trabalhadores urbanos e rurais, encontramos no art. 7º da CF, reconhecimentos de direitos tais como a licença à gestante, sem prejuízo do emprego e do salário, com a duração de cento e vinte dias e, licença paternidade, nos termos fixados em lei. A motivação para a apresentação deste projeto de lei está circunstanciada em fatos divulgados, recentemente pela mídia Correio Web 05 09 2001 Homem ganha um mês de licença paternidade para conviver com o filho adotivo. Um telefonema da Vara de Infância de Brasília uniu os destinos do geógrafo Otaviano Batista, 52 anos e do menino João Carlos. Solteiro, Otaviano queria encarar sozinho o desafio da paternidade. Como a maioria das pessoas que procuram a adoção, o pedido era por um menino branco e com menos de dois anos, mas o que lhe foi apresentado pela assistente social foi esse garoto de oito anos. Após o encontro, a simpatia venceu os receios e, há um ano, os dois vivem juntos. Ora como a Consolidação das Leis do Trabalho garante à mãe adotiva o direito à licença maternidade, esse cidadão entendeu que o CÂMARA DOS DEPUTADOS dispositivo legal previsto na CLT, abriria o precedente para a situação de um pai adotivo e solteiro. Utilizando se do Estatuto da Criança e do Adolescente, principalmente por conta do estatuído no art. 227. Otaviano é funcionário do Conselho Federal de Engenharia, Arquitetura e Agronomia Confea há 33 anos. No mesmo dia em que conseguiu a nova certidão de nascimento de João Carlos, o geógrafo protocolou o pedido de licença paternidade. A seção de Recursos Humanos pediu auxílio para o Departamento Jurídico, que elaborou um parecer favorável à concessão dos 120 dias. O documento foi submetido à votação dos conselheiros e eles concederam a licença ao pai adotivo. Mas três dias depois, o conselho voltou atrás e admitiu, excepcionalmente, 30 dias consecutivos de licença paternidade já que não existe fundamento legal para conceder a licença paternidade pelo prazo de 120 dias Ainda que não haja uma lei dizendo especificamente que o pai adotivo solteiro tem direito a 120 dias de licença, a legislação precisa estar a serviço da Justiça, não da burocracia completa o procurador do Confea. Outro exemplo é uma decisão inédita, ocorrida no último dia 22 de agosto, proferida pelo Tribunal de Justiça de Pernambuco, que concedeu a primeira licença paternidade de 180 dias para um servidor do Poder Judiciário de Pernambuco, que adotou uma criança de quatro meses. A decisão destaca a importância da convivência e alega que, quando um pai solteiro adota, ele é pai e mãe. Por isso, precisa se dedicar ao filho. Quando uma criança é adotada em idade tão delicada, precisa de atenção especial nos primeiros meses de convivência. Esse acompanhamento, afetivo e efetivo, vai ser determinante para toda a sua história diz a decisão. Para a psicóloga Soraya Pereira, presidente da ONG Aconchego, o momento de convívio entre pai e filho é essencial para o sucesso de uma adoção tardia. As duas histórias se encontraram agora e eles precisam adquirir cumplicidade. Afinal, é uma família que está se construindo. Esse tempo junto é preciso defende. Destarte que, hoje, a Consolidação das Leis do Trabalho, aprovada pelo Decreto Lei nº 5. 452, de 1º de maio de 1943, em seu artigo 392 A, assim preconiza A empregada que adotar ou obtiver guarda judicial para fins de adoção de criança será concedida licença maternidade nos termos do art. 392, observado o disposto no seu 5º. Já, a Lei nº 8. 213, de 24 de julho de 1991, em seu artigo 71 A, assim estabelece A segurada da Previdência Social que adotar ou obtiver guarda judicial para fins de adoção de criança é devido salário maternidade pelo período de 120 cento e vinte dias, se a criança tiver até 1 um ano de idade, de 60 sessenta dias, se a CÂMARA DOS DEPUTADOS criança tiver entre 1 um e 4 quatro anos de idade, e 30 trinta dias, se a criança tiver de 4 quatro a 8 oito anos de idade. Por conclusão, à vista de tudo aqui exposado em relação à possibilidade de estarmos aprovando mais um procedimento isonômico, muito mais com a preocupação parlamentar de se fazer prevalecer os direitos constitucionais já definidos pela Carta Magna, são os motivos mais que bastantes para que possa propor a aprovação deste Projeto de Lei. Sala das Sessões, em de setembro de 2011. Deputada Andreia Zito PSDB RJ</code> | <code>1</code> | * Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters: ```json { "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", "margin": 0.5, "size_average": true } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `num_train_epochs`: 1 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters <details><summary>Click to expand</summary> - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: no - `prediction_loss_only`: True - `per_device_train_batch_size`: 2 - `per_device_eval_batch_size`: 2 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `restore_callback_states_from_checkpoint`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `eval_do_concat_batches`: True - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_eval_metrics`: False - `eval_on_start`: False - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin </details> ### Training Logs | Epoch | Step | Training Loss | |:------:|:----:|:-------------:| | 0.2311 | 500 | 0.0242 | | 0.4621 | 1000 | 0.0048 | | 0.6932 | 1500 | 0.0036 | | 0.9242 | 2000 | 0.0038 | ### Framework Versions - Python: 3.10.12 - Sentence Transformers: 3.0.1 - Transformers: 4.42.4 - PyTorch: 2.3.1+cu121 - Accelerate: 0.32.1 - Datasets: 2.20.0 - Tokenizers: 0.19.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### ContrastiveLoss ```bibtex @inproceedings{hadsell2006dimensionality, author={Hadsell, R. and Chopra, S. and LeCun, Y.}, booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)}, title={Dimensionality Reduction by Learning an Invariant Mapping}, year={2006}, volume={2}, number={}, pages={1735-1742}, doi={10.1109/CVPR.2006.100} } ``` <!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->