Split (1)

train · 38.5k rows

modelId stringlengths 4 112	sha stringlengths 40 40	lastModified stringlengths 24 24	tags sequence	pipeline_tag stringclasses 29 values	private bool 1 class	author stringlengths 2 38 ⌀	config null	id stringlengths 4 112	downloads float64 0 36.8M ⌀	likes float64 0 712 ⌀	library_name stringclasses 17 values	__index_level_0__ int64 0 38.5k	readme stringlengths 0 186k
hackathon-pln-es/t5-small-finetuned-spanish-to-quechua	1a1c1a24c23a9b4cff2444718dcbc958875018cb	2022-04-03T05:42:56.000Z	[ "pytorch", "t5", "text2text-generation", "es", "qu", "transformers", "quechua", "translation", "spanish", "license:apache-2.0", "autotrain_compatible" ]	translation	false	hackathon-pln-es	null	hackathon-pln-es/t5-small-finetuned-spanish-to-quechua	90	4	transformers	4,800	--- language: - es - qu tags: - quechua - translation - spanish license: apache-2.0 metrics: - bleu - sacrebleu widget: - text: "Dios ama a los hombres" - text: "A pesar de todo, soy feliz" - text: "¿Qué harán allí?" - text: "Debes aprender a respetar" --- # Spanish to Quechua translator This model is a finetuned version of the [t5-small](https://huggingface.co/t5-small). ## Model description t5-small-finetuned-spanish-to-quechua has trained for 46 epochs with 102 747 sentences, the validation was performed with 12 844 sentences and 12 843 sentences were used for the test. ## Intended uses & limitations A large part of the dataset has been extracted from biblical texts, which makes the model perform better with certain types of sentences. ### How to use You can import this model as follows: ```python >>> from transformers import AutoModelForSeq2SeqLM, AutoTokenizer >>> model_name = 'hackathon-pln-es/t5-small-finetuned-spanish-to-quechua' >>> model = AutoModelForSeq2SeqLM.from_pretrained(model_name) >>> tokenizer = AutoTokenizer.from_pretrained(model_name) ``` To translate you can do: ```python >>> sentence = "Entonces dijo" >>> input = tokenizer(sentence, return_tensors="pt") >>> output = model.generate(input["input_ids"], max_length=40, num_beams=4, early_stopping=True) >>> print('Original Sentence: {} \nTranslated sentence: {}'.format(sentence, tokenizer.decode(output[0]))) ``` ### Limitations and bias Actually this model only can translate to Quechua of Ayacucho. ## Training data For train this model we use [Spanish to Quechua dataset](https://huggingface.co/datasets/hackathon-pln-es/spanish-to-quechua) ## Evaluation results We obtained the following metrics during the training process: - eval_bleu = 2.9691 - eval_loss = 1.2064628601074219 ## Team members - [Sara Benel](https://huggingface.co/sbenel) - [Jose Vílchez](https://huggingface.co/JCarlos)
ccdv/lsg-bart-base-16384-pubmed	ee321b6636965d14801b58b61362a34b5704a23c	2022-07-25T05:29:10.000Z	[ "pytorch", "bart", "text2text-generation", "en", "dataset:scientific_papers", "transformers", "summarization", "model-index", "autotrain_compatible" ]	summarization	false	ccdv	null	ccdv/lsg-bart-base-16384-pubmed	90	3	transformers	4,801	--- language: - en tags: - summarization datasets: - scientific_papers metrics: - rouge model-index: - name: ccdv/lsg-bart-base-16384-pubmed 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. --> This model relies on a custom modeling file, you need to add trust_remote_code=True\ See [\#13467](https://github.com/huggingface/transformers/pull/13467) ```python from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline tokenizer = AutoTokenizer.from_pretrained("ccdv/lsg-bart-base-16384-pubmed", trust_remote_code=True) model = AutoModelForSeq2SeqLM.from_pretrained("ccdv/lsg-bart-base-16384-pubmed", trust_remote_code=True) text = "Replace by what you want." pipe = pipeline("text2text-generation", model=model, tokenizer=tokenizer, device=0) generated_text = pipe( text, truncation=True, max_length=64, no_repeat_ngram_size=7, num_beams=2, early_stopping=True ) ``` # ccdv/lsg-bart-base-16384-pubmed This model is a fine-tuned version of [ccdv/lsg-bart-base-4096-pubmed](https://huggingface.co/ccdv/lsg-bart-base-4096-pubmed) on the [scientific_papers pubmed](https://huggingface.co/datasets/scientific_papers) dataset. \ The model is converted to handle 16384 long sequences and fine-tuned accordingly during 1 epoch. \ It achieves the following results on the test set: \| Length \| Global tokens \| Fine-tuning \| Block Size \| Sparsity \| Connexions \| R1 \| R2 \| RL \| RLsum \| \|:------ \|:------------- \|:----------- \|:---------- \|:-------- \| :--------- \|:----- \|:----- \|:----- \|:----- \| \| 16384 \| 64 \| Full \| 256 \| 0 \| 768 \| 48.32 \| 22.52 \| 29.36 \| 44.57 \| \| 16384 \| 1 \| Full \| 256 \| 0 \| 768 \| 48.26 \| 22.53 \| 29.40 \| 44.51 \| \| 16384 \| 64 \| Global only \| 256 \| 0 \| 768 \| 48.12 \| 20.46 \| 29.34 \| 44.40 \| \| 16384 \| 1 \| None \| 256 \| 0 \| 768 \| 48.03 \| 22.42 \| 29.28 \| 44.32 \| Reference model: \| Length \| Global tokens \| Fine-tuning \| Block Size \| Sparsity \| Connexions \| R1 \| R2 \| RL \| RLsum \| \|:------ \|:------------- \|:----------- \|:---------- \|:-------- \| :--------- \|:----- \|:----- \|:----- \|:----- \| \| 4096 \| 1 \| - \| 256 \| 0 \| 768 \| 47.37 \| 21.74 \| 28.59 \| 43.67 \| ## Model description The model relies on Local-Sparse-Global attention to handle long sequences: ![attn](attn.png) The model has about ~145 millions parameters (6 encoder layers - 6 decoder layers). \ The model is warm started from [ccdv/lsg-bart-base-4096-pubmed](https://huggingface.co/ccdv/lsg-bart-base-4096-pubmed), converted to handle long sequences (encoder only) and fine tuned. ## 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: 8e-05 - train_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 1.0 ### Generate hyperparameters The following hyperparameters were used during generation: - dataset_name: scientific_papers - dataset_config_name: pubmed - eval_batch_size: 4 - eval_samples: 6658 - early_stopping: True - ignore_pad_token_for_loss: True - length_penalty: 2.0 - max_length: 512 - min_length: 128 - num_beams: 5 - no_repeat_ngram_size: None - seed: 123 ### Framework versions - Transformers 4.18.0 - Pytorch 1.10.1+cu102 - Datasets 2.1.0 - Tokenizers 0.11.6
sanjay-m1/parrot-adequacy-on-BART	0514f368694920fc841582af98ed94f720a00df2	2022-05-21T17:37:29.000Z	[ "pytorch", "bart", "transformers" ]	null	false	sanjay-m1	null	sanjay-m1/parrot-adequacy-on-BART	90	null	transformers	4,802	# Parrot THIS IS AN ANCILLARY MODEL FOR PARROT PARAPHRASER ## 1. What is Parrot? Parrot is a paraphrase based utterance augmentation framework purpose built to accelerate training NLU models. A paraphrase framework is more than just a paraphrasing model. Please refer to the [github page](https://github.com/PrithivirajDamodaran/Parrot) or The model card prithivida/parrot_paraphraser_on_T5
eunsour/en-ko-transliterator	169f2b5772f9347985d0a7ab462685c801b8cbfe	2022-06-28T14:32:25.000Z	[ "pytorch", "mt5", "text2text-generation", "transformers", "autotrain_compatible" ]	text2text-generation	false	eunsour	null	eunsour/en-ko-transliterator	90	0	transformers	4,803	``` !pip install simpletransformers from simpletransformers.t5 import T5Model model = T5Model("mt5", "eunsour/en-ko-transliterator", use_cuda=False) print(model.predict(["transformer"])) print(model.predict(["attention"])) ```
JulesBelveze/t5-small-headline-generator	58ef632459769780866fa6d16a9cfbc69add8cb0	2022-07-01T05:17:57.000Z	[ "pytorch", "t5", "text2text-generation", "en", "dataset:JulesBelveze/tldr_news", "transformers", "summarization", "headline-generation", "text-generation", "autotrain_compatible" ]	summarization	false	JulesBelveze	null	JulesBelveze/t5-small-headline-generator	90	2	transformers	4,804	--- language: - en tags: - summarization - headline-generation - text-generation datasets: - JulesBelveze/tldr_news metrics: - rouge1 - rouge2 - rougeL - rougeLsum --- # t5-small for headline generation This model is a [t5-small](https://huggingface.co/t5-small) fine-tuned for headline generation using the [JulesBelveze/tldr_news](https://huggingface.co/datasets/JulesBelveze/tldr_news) dataset. ## Using this model ```python import re from transformers import AutoTokenizer, T5ForConditionalGeneration WHITESPACE_HANDLER = lambda k: re.sub('\s+', ' ', re.sub('\n+', ' ', k.strip())) article_text = """US FCC commissioner Brendan Carr has asked Apple and Google to remove TikTok from their app stores. The video app is owned by Chinese company ByteDance. Carr claims that TikTok functions as a surveillance tool that harvests extensive amounts of personal and sensitive data from US citizens. TikTok says its data access approval process is overseen by a US-based security team and that data is only accessed on an as-needed basis under strict controls.""" model_name = "JulesBelveze/t5-small-headline-generator" tokenizer = AutoTokenizer.from_pretrained(model_name) model = T5ForConditionalGeneration.from_pretrained(model_name) input_ids = tokenizer( [WHITESPACE_HANDLER(article_text)], return_tensors="pt", padding="max_length", truncation=True, max_length=384 )["input_ids"] output_ids = model.generate( input_ids=input_ids, max_length=84, no_repeat_ngram_size=2, num_beams=4 )[0] summary = tokenizer.decode( output_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False ) print(summary) ``` ## Evaluation \| Metric \| Score \| \|------------\|---------\| \| ROUGE 1 \| 44.2379 \| \| ROUGE 2 \| 17.4961 \| \| ROUGE L \| 41.1119 \| \| ROUGE Lsum \| 41.1256 \|
Amrrs/indian-foods	f17394f1fb1e91fc497c5124805a0ca7db91dc03	2021-07-20T10:20:55.000Z	[ "pytorch", "tensorboard", "vit", "image-classification", "transformers", "huggingpics", "model-index" ]	image-classification	false	Amrrs	null	Amrrs/indian-foods	89	1	transformers	4,805	--- tags: - image-classification - pytorch - huggingpics metrics: - accuracy model-index: - name: indian-foods results: - task: name: Image Classification type: image-classification metrics: - name: Accuracy type: accuracy value: 0.9285714030265808 --- # indian-foods Autogenerated by HuggingPics🤗🖼️ Create your own image classifier for anything by running [the demo on Google Colab](https://colab.research.google.com/github/nateraw/huggingpics/blob/main/HuggingPics.ipynb). Report any issues with the demo at the [github repo](https://github.com/nateraw/huggingpics). ## Example Images #### idli ![idli](images/idli.jpg) #### kachori ![kachori](images/kachori.jpg) #### pani puri ![pani puri](images/pani_puri.jpg) #### samosa ![samosa](images/samosa.jpg) #### vada pav ![vada pav](images/vada_pav.jpg)
DeepChem/SmilesTokenizer_PubChem_1M	4922f620c0488ad0362ba89430f8c96041072c29	2021-05-31T20:54:05.000Z	[ "pytorch", "roberta", "feature-extraction", "transformers" ]	feature-extraction	false	DeepChem	null	DeepChem/SmilesTokenizer_PubChem_1M	89	null	transformers	4,806	RoBERTa model trained on 1M SMILES from PubChem 77M set in MoleculeNet. Uses Smiles-Tokenizer
Helsinki-NLP/opus-mt-de-ZH	93d4bc065a572a35ab1f1110ffeccc9740444a42	2021-09-09T21:30:07.000Z	[ "pytorch", "marian", "text2text-generation", "de", "zh", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-de-ZH	89	1	transformers	4,807	--- tags: - translation license: apache-2.0 --- ### opus-mt-de-ZH * source languages: de * target languages: cmn,cn,yue,ze_zh,zh_cn,zh_CN,zh_HK,zh_tw,zh_TW,zh_yue,zhs,zht,zh * OPUS readme: [de-cmn+cn+yue+ze_zh+zh_cn+zh_CN+zh_HK+zh_tw+zh_TW+zh_yue+zhs+zht+zh](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/de-cmn+cn+yue+ze_zh+zh_cn+zh_CN+zh_HK+zh_tw+zh_TW+zh_yue+zhs+zht+zh/README.md) * dataset: opus * model: transformer-align * pre-processing: normalization + SentencePiece * a sentence initial language token is required in the form of `>>id<<` (id = valid target language ID) * download original weights: [opus-2020-01-20.zip](https://object.pouta.csc.fi/OPUS-MT-models/de-cmn+cn+yue+ze_zh+zh_cn+zh_CN+zh_HK+zh_tw+zh_TW+zh_yue+zhs+zht+zh/opus-2020-01-20.zip) * test set translations: [opus-2020-01-20.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/de-cmn+cn+yue+ze_zh+zh_cn+zh_CN+zh_HK+zh_tw+zh_TW+zh_yue+zhs+zht+zh/opus-2020-01-20.test.txt) * test set scores: [opus-2020-01-20.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/de-cmn+cn+yue+ze_zh+zh_cn+zh_CN+zh_HK+zh_tw+zh_TW+zh_yue+zhs+zht+zh/opus-2020-01-20.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| bible-uedin.de.zh \| 24.4 \| 0.335 \|
Helsinki-NLP/opus-mt-es-vi	b3b69971697a56c967a303713e0f4dcd06256311	2021-01-18T08:29:49.000Z	[ "pytorch", "marian", "text2text-generation", "es", "vi", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-es-vi	89	null	transformers	4,808	--- language: - es - vi tags: - translation license: apache-2.0 --- ### spa-vie * source group: Spanish * target group: Vietnamese * OPUS readme: [spa-vie](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-vie/README.md) * model: transformer-align * source language(s): spa * target language(s): vie * model: transformer-align * pre-processing: normalization + SentencePiece (spm32k,spm32k) * download original weights: [opus-2020-06-17.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-vie/opus-2020-06-17.zip) * test set translations: [opus-2020-06-17.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-vie/opus-2020-06-17.test.txt) * test set scores: [opus-2020-06-17.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/spa-vie/opus-2020-06-17.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba-test.spa.vie \| 33.1 \| 0.508 \| ### System Info: - hf_name: spa-vie - source_languages: spa - target_languages: vie - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-vie/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['es', 'vi'] - src_constituents: {'spa'} - tgt_constituents: {'vie', 'vie_Hani'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/spa-vie/opus-2020-06-17.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/spa-vie/opus-2020-06-17.test.txt - src_alpha3: spa - tgt_alpha3: vie - short_pair: es-vi - chrF2_score: 0.508 - bleu: 33.1 - brevity_penalty: 0.98 - ref_len: 4654.0 - src_name: Spanish - tgt_name: Vietnamese - train_date: 2020-06-17 - src_alpha2: es - tgt_alpha2: vi - prefer_old: False - long_pair: spa-vie - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
MKaan/multilingual-cpv-sector-classifier	593b7b5901a38e28cdf27fe3381ae96dc172415e	2021-11-28T13:09:32.000Z	[ "pytorch", "bert", "text-classification", "transformers", "eu", "public procurement", "cpv", "sector", "multilingual", "license:apache-2.0" ]	text-classification	false	MKaan	null	MKaan/multilingual-cpv-sector-classifier	89	3	transformers	4,809	--- license: apache-2.0 tags: - eu - public procurement - cpv - sector - multilingual - transformers - text-classification widget: - text: "Oppegård municipality, hereafter called the contracting authority, intends to enter into a framework agreement with one supplier for the procurement of fresh bread and bakery products for Oppegård municipality. The contract is estimated to NOK 1 400 000 per annum excluding VAT The total for the entire period including options is NOK 5 600 000 excluding VAT" --- # multilingual-cpv-sector-classifier This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on [the Tenders Economic Daily Public Procurement Data](https://simap.ted.europa.eu/en). It achieves the following results on the evaluation set: - F1 Score: 0.686 ## Model description The model takes procurement descriptions written in any of [104 languages](https://github.com/google-research/bert/blob/master/multilingual.md#list-of-languages) and classifies them into 45 sector classes represented by [CPV(Common Procurement Vocabulary)](https://simap.ted.europa.eu/en_GB/web/simap/cpv) code descriptions as listed below. \| Common Procurement Vocabulary \| \|:-----------------------------\| \| Administration, defence and social security services. 👮‍♀️ \| \| Agricultural machinery. 🚜 \| \| Agricultural, farming, fishing, forestry and related products. 🌾 \| \| Agricultural, forestry, horticultural, aquacultural and apicultural services. 👨🏿‍🌾 \| \| Architectural, construction, engineering and inspection services. 👷‍♂️ \| \| Business services: law, marketing, consulting, recruitment, printing and security. 👩‍💼 \| \| Chemical products. 🧪 \| \| Clothing, footwear, luggage articles and accessories. 👖 \| \| Collected and purified water. 🌊 \| \| Construction structures and materials; auxiliary products to construction (excepts electric apparatus). 🧱 \| \| Construction work. 🏗️ \| \| Education and training services. 👩🏿‍🏫 \| \| Electrical machinery, apparatus, equipment and consumables; Lighting. ⚡ \| \| Financial and insurance services. 👨‍💼 \| \| Food, beverages, tobacco and related products. 🍽️ \| \| Furniture (incl. office furniture), furnishings, domestic appliances (excl. lighting) and cleaning products. 🗄️ \| \| Health and social work services. 👨🏽‍⚕️ \| \| Hotel, restaurant and retail trade services. 🏨 \| \| IT services: consulting, software development, Internet and support. 🖥️ \| \| Industrial machinery. 🏭 \| \| Installation services (except software). 🛠️ \| \| Laboratory, optical and precision equipments (excl. glasses). 🔬 \| \| Leather and textile fabrics, plastic and rubber materials. 🧵 \| \| Machinery for mining, quarrying, construction equipment. ⛏️ \| \| Medical equipments, pharmaceuticals and personal care products. 💉 \| \| Mining, basic metals and related products. ⚙️ \| \| Musical instruments, sport goods, games, toys, handicraft, art materials and accessories. 🎸 \| \| Office and computing machinery, equipment and supplies except furniture and software packages. 🖨️ \| \| Other community, social and personal services. 🧑🏽‍🤝‍🧑🏽 \| \| Petroleum products, fuel, electricity and other sources of energy. 🔋 \| \| Postal and telecommunications services. 📶 \| \| Printed matter and related products. 📰 \| \| Public utilities. ⛲ \| \| Radio, television, communication, telecommunication and related equipment. 📡 \| \| Real estate services. 🏠 \| \| Recreational, cultural and sporting services. 🚴 \| \| Repair and maintenance services. 🔧 \| \| Research and development services and related consultancy services. 👩‍🔬 \| \| Security, fire-fighting, police and defence equipment. 🧯 \| \| Services related to the oil and gas industry. ⛽ \| \| Sewage-, refuse-, cleaning-, and environmental services. 🧹 \| \| Software package and information systems. 🔣 \| \| Supporting and auxiliary transport services; travel agencies services. 🚃 \| \| Transport equipment and auxiliary products to transportation. 🚌 \| \| Transport services (excl. Waste transport). 💺 ## Intended uses & limitations - Input description should be written in any of [the 104 languages](https://github.com/google-research/bert/blob/master/multilingual.md#list-of-languages) that MBERT supports. - The model is just evaluated in 22 languages. Thus there is no information about the performances in other languages. - The domain is also restricted by the awarded procurement notice descriptions in European Union. Evaluating on whole document texts might change the performance. ## Training and evaluation data - The whole data consists of 744,360 rows. Shuffled and split into train and validation sets by using 80%/20% manner. - Each description represents a unique contract notice description awarded between 2011 and 2018. - Both training and validation data have contract notice descriptions written in 22 European Languages. (Malta and Irish are extracted due to scarcity compared to whole data) ## Training procedure The training procedure has been completed on Google Cloud V3-8 TPUs. Thanks [Google](https://sites.research.google/trc/about/) for giving the access to Cloud TPUs ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - num_epochs: 3 - gradient_accumulation_steps: 8 - batch_size_per_device: 4 - total_train_batch_size: 32 ### Training results \| Epoch \| Step \| F1 Score\| \|:-----:\|:------:\|:------:\| \| 1 \| 18,609 \| 0.630 \| \| 2 \| 37,218 \| 0.674 \| \| 3 \| 55,827 \| 0.686 \| \| Language\| F1 Score\| Test Size\| \|:-----:\|:-----:\|:-----:\| \| PL\| 0.759\| 13950\| \| RO\| 0.736\| 3522\| \| SK\| 0.719\| 1122\| \| LT\| 0.687\| 2424\| \| HU\| 0.681\| 1879\| \| BG\| 0.675\| 2459\| \| CS\| 0.668\| 2694\| \| LV\| 0.664\| 836\| \| DE\| 0.645\| 35354\| \| FI\| 0.644\| 1898\| \| ES\| 0.643\| 7483\| \| PT\| 0.631\| 874\| \| EN\| 0.631\| 16615\| \| HR\| 0.626\| 865\| \| IT\| 0.626\| 8035\| \| NL\| 0.624\| 5640\| \| EL\| 0.623\| 1724\| \| SL\| 0.615\| 482\| \| SV\| 0.607\| 3326\| \| DA\| 0.603\| 1925\| \| FR\| 0.601\| 33113\| \| ET\| 0.572\| 458\|\|
NDugar/v3-Large-mnli	291554748314581adf4398af85d52b5f14f7f70e	2021-12-26T15:27:41.000Z	[ "pytorch", "deberta-v2", "text-classification", "en", "transformers", "deberta-v1", "deberta-mnli", "license:mit", "zero-shot-classification" ]	zero-shot-classification	false	NDugar	null	NDugar/v3-Large-mnli	89	1	transformers	4,810	--- language: en tags: - deberta-v1 - deberta-mnli tasks: mnli thumbnail: https://huggingface.co/front/thumbnails/microsoft.png license: mit pipeline_tag: zero-shot-classification --- This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co/microsoft/deberta-v3-large) on the GLUE MNLI dataset. It achieves the following results on the evaluation set: - Loss: 0.4103 - Accuracy: 0.9175 ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 6e-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 - lr_scheduler_warmup_steps: 50 - num_epochs: 2.0 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| \|:-------------:\|:-----:\|:-----:\|:---------------:\|:--------:\| \| 0.3631 \| 1.0 \| 49088 \| 0.3129 \| 0.9130 \| \| 0.2267 \| 2.0 \| 98176 \| 0.4157 \| 0.9153 \| ### Framework versions - Transformers 4.13.0.dev0 - Pytorch 1.10.0 - Datasets 1.15.2.dev0 - Tokenizers 0.10.3
PaulLerner/dpr_context_encoder_triviaqa_without_viquae	d1c83af9b648a6dbe234dfb51c4713ecceed2d9f	2022-02-18T13:58:18.000Z	[ "pytorch", "dpr", "transformers" ]	null	false	PaulLerner	null	PaulLerner/dpr_context_encoder_triviaqa_without_viquae	89	null	transformers	4,811	Entry not found
cosmoquester/bart-ko-small	32c6ec5844197a774d6460e7fc83812af9dc24e3	2021-08-28T05:09:54.000Z	[ "pytorch", "tf", "bart", "text2text-generation", "ko", "transformers", "autotrain_compatible" ]	text2text-generation	false	cosmoquester	null	cosmoquester/bart-ko-small	89	null	transformers	4,812	--- language: ko --- # Pretrained BART in Korean This is pretrained BART model with multiple Korean Datasets. I used multiple datasets for generalizing the model for both colloquial and written texts. The training is supported by [TPU Research Cloud](https://sites.research.google/trc/) program. The script which is used to pre-train model is [here](https://github.com/cosmoquester/transformers-bart-pretrain). When you use the reference API, you must wrap the sentence with `[BOS]` and `[EOS]` like below example. ``` [BOS] 안녕하세요? 반가워요~~ [EOS] ``` You can also test mask filling performance using `[MASK]` token like this. ``` [BOS] [MASK] 먹었어? [EOS] ``` ## Benchmark <style> table { border-collapse: collapse; border-style: hidden; width: 100%; } td, th { border: 1px solid #4d5562; padding: 8px; } </style> <table> <tr> <th>Dataset</th> <td>KLUE NLI dev</th> <td>NSMC test</td> <td>QuestionPair test</td> <td colspan="2">KLUE TC dev</td> <td colspan="3">KLUE STS dev</td> <td colspan="3">KorSTS dev</td> <td colspan="2">HateSpeech dev</td> </tr> <tr> <th>Metric</th> <!-- KLUE NLI --> <td>Acc</th> <!-- NSMC --> <td>Acc</td> <!-- QuestionPair --> <td>Acc</td> <!-- KLUE TC --> <td>Acc</td> <td>F1</td> <!-- KLUE STS --> <td>F1</td> <td>Pearson</td> <td>Spearman</td> <!-- KorSTS --> <td>F1</td> <td>Pearson</td> <td>Spearman</td> <!-- HateSpeech --> <td>Bias Acc</td> <td>Hate Acc</td> </tr> <tr> <th>Score</th> <!-- KLUE NLI --> <td>0.639</th> <!-- NSMC --> <td>0.8721</td> <!-- QuestionPair --> <td>0.905</td> <!-- KLUE TC --> <td>0.8551</td> <td>0.8515</td> <!-- KLUE STS --> <td>0.7406</td> <td>0.7593</td> <td>0.7551</td> <!-- KorSTS --> <td>0.7897</td> <td>0.7269</td> <td>0.7037</td> <!-- HateSpeech --> <td>0.8068</td> <td>0.5966</td> </tr> </table> - The performance was measured using [the notebooks here](https://github.com/cosmoquester/transformers-bart-finetune) with colab. ## Used Datasets ### [모두의 말뭉치](https://corpus.korean.go.kr/) - 일상 대화 말뭉치 2020 - 구어 말뭉치 - 문어 말뭉치 - 신문 말뭉치 ### AIhub - [개방데이터 전문분야말뭉치](https://aihub.or.kr/aidata/30717) - [개방데이터 한국어대화요약](https://aihub.or.kr/aidata/30714) - [개방데이터 감성 대화 말뭉치](https://aihub.or.kr/aidata/7978) - [개방데이터 한국어 음성](https://aihub.or.kr/aidata/105) - [개방데이터 한국어 SNS](https://aihub.or.kr/aidata/30718) ### [세종 말뭉치](https://ithub.korean.go.kr/)
mrm8488/deberta-v3-small-finetuned-sst2	94d075b4095d80dde2c23ad513ebcac7bfe7f0ad	2021-11-21T19:17:56.000Z	[ "pytorch", "tensorboard", "deberta-v2", "text-classification", "en", "dataset:glue", "transformers", "generated_from_trainer", "deberta-v3", "license:mit", "model-index" ]	text-classification	false	mrm8488	null	mrm8488/deberta-v3-small-finetuned-sst2	89	null	transformers	4,813	--- language: - en license: mit tags: - generated_from_trainer - deberta-v3 datasets: - glue metrics: - accuracy model-index: - name: deberta-v3-small results: - task: name: Text Classification type: text-classification dataset: name: GLUE SST2 type: glue args: sst2 metrics: - name: Accuracy type: accuracy value: 0.9403669724770642 --- <!-- 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. --> # DeBERTa v3 (small) fine-tuned on SST2 This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the GLUE SST2 dataset. It achieves the following results on the evaluation set: - Loss: 0.2134 - Accuracy: 0.9404 ## 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-05 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5.0 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| \|:-------------:\|:-----:\|:-----:\|:---------------:\|:--------:\| \| 0.176 \| 1.0 \| 4210 \| 0.2134 \| 0.9404 \| \| 0.1254 \| 2.0 \| 8420 \| 0.2362 \| 0.9415 \| \| 0.0957 \| 3.0 \| 12630 \| 0.3187 \| 0.9335 \| \| 0.0673 \| 4.0 \| 16840 \| 0.3039 \| 0.9266 \| \| 0.0457 \| 5.0 \| 21050 \| 0.3521 \| 0.9312 \| ### Framework versions - Transformers 4.13.0.dev0 - Pytorch 1.10.0+cu111 - Datasets 1.15.1 - Tokenizers 0.10.3
salesken/paraphrase_generation	1141e8ba54581a8130c4adcc407ae85525d1cf4f	2021-05-23T12:33:04.000Z	[ "pytorch", "jax", "gpt2", "text-generation", "en", "transformers", "salesken", "license:apache-2.0" ]	text-generation	false	salesken	null	salesken/paraphrase_generation	89	1	transformers	4,814	--- language: en thumbnail: https://salesken.ai/assets/images/logo.png license: apache-2.0 inference: false widget: - text: "every moment is a fresh beginning" tags: salesken --- Use this model to generate variations to augment the training data used for NLU systems. ```python from transformers import AutoTokenizer, AutoModelWithLMHead import torch if torch.cuda.is_available(): device = torch.device("cuda") else : device = "cpu" tokenizer = AutoTokenizer.from_pretrained("salesken/paraphrase_generation") model = AutoModelWithLMHead.from_pretrained("salesken/paraphrase_generation").to(device) input_query="every moment is a fresh beginning" query= input_query + " ~~ " input_ids = tokenizer.encode(query.lower(), return_tensors='pt').to(device) sample_outputs = model.generate(input_ids, do_sample=True, num_beams=1, max_length=128, temperature=0.9, top_p= 0.99, top_k = 30, num_return_sequences=40) paraphrases = [] for i in range(len(sample_outputs)): r = tokenizer.decode(sample_outputs[i], skip_special_tokens=True).split('\|\|')[0] r = r.split(' ~~ ')[1] if r not in paraphrases: paraphrases.append(r) print(paraphrases) ``` To evaluate if a paraphrase is a semantic variation to the input query or just a surface level variation & rank the generated paraphrases, use the following model: https://huggingface.co/salesken/paraphrase_diversity_ranker
timm/vit_huge_patch14_224_in21k	bbf572e74b9cf2a10daa9461505cb506c9189c4a	2021-03-18T10:58:13.000Z	[ "pytorch", "dataset:imagenet_21k", "timm", "image-classification", "vision-transformer", "license:apache-2.0" ]	image-classification	false	timm	null	timm/vit_huge_patch14_224_in21k	89	null	timm	4,815	--- tags: - image-classification - timm - vision-transformer license: apache-2.0 datasets: - imagenet_21k inference: false --- # ViT-H/14 (ImageNet-21k) ...
yongzx/gpt2-finetuned-oscar-fr-ori-tok	b3f3f7e2dc1729199e44424c6b7edc1aaafa18f6	2021-12-09T06:37:06.000Z	[ "pytorch", "gpt2", "text-generation", "fr", "dataset:oscar", "transformers", "license:mit" ]	text-generation	false	yongzx	null	yongzx/gpt2-finetuned-oscar-fr-ori-tok	89	null	transformers	4,816	--- language: - fr tags: - text-generation license: mit datasets: - oscar widget: - text: "Je suis ravi de vous " --- # GPT-2 finetuned on French Dataset ### Tokenizer We use GPT-2 tokenizer. ### Model We finetuned the `wte` and `wpe` layers of GPT-2 (while freezing the parameters of all other layers) on OSCAR's `unshuffled_original_fr` French data subset. We used [Huggingface's code](https://github.com/huggingface/transformers/blob/master/examples/pytorch/language-modeling/run_clm.py) for fine-tuning the causal language model GPT-2, but with the following parameters changed ``` - preprocessing_num_workers: 8 - per_device_train_batch_size: 2 - gradient_accumulation_steps: 4 - per_device_eval_batch_size: 2 - eval_accumulation_steps: 4 - eval_steps: 1000 - evaluation_strategy: "steps" - max_eval_samples: 5000 ``` Final checkpoint: checkpoint-76500
ali2066/bert-base-uncased_token_itr0_0.0001_TRAIN_essays_TEST_editorials_05_03_2022-06_21_38	328afb20b84511cb9c350d495c6a3a90922aefc1	2022-03-05T05:24:04.000Z	[ "pytorch", "tensorboard", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	ali2066	null	ali2066/bert-base-uncased_token_itr0_0.0001_TRAIN_essays_TEST_editorials_05_03_2022-06_21_38	89	null	transformers	4,817	Entry not found
jkhan447/sentiment-model-sample-27go-emotion	2e63119a9d2b37e5e054e903f62cded65ee34a70	2022-04-01T08:13:56.000Z	[ "pytorch", "tensorboard", "bert", "text-classification", "dataset:go_emotions", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]	text-classification	false	jkhan447	null	jkhan447/sentiment-model-sample-27go-emotion	89	null	transformers	4,818	--- license: apache-2.0 tags: - generated_from_trainer datasets: - go_emotions metrics: - accuracy model-index: - name: sentiment-model-sample-27go-emotion results: - task: name: Text Classification type: text-classification dataset: name: go_emotions type: go_emotions args: simplified metrics: - name: Accuracy type: accuracy value: 0.5888888888888889 --- <!-- 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. --> # sentiment-model-sample-27go-emotion This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the go_emotions dataset. It achieves the following results on the evaluation set: - Loss: 4.1765 - Accuracy: 0.5889 ## 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: 32 - eval_batch_size: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 50 ### Training results ### Framework versions - Transformers 4.17.0 - Pytorch 1.10.0+cu111 - Datasets 2.0.0 - Tokenizers 0.12.0
ismail-lucifer011/autotrain-company_all-903429548	acbdc6779e48dfde49f2ba77d6f2cfb431333d16	2022-05-24T14:24:20.000Z	[ "pytorch", "distilbert", "token-classification", "en", "dataset:ismail-lucifer011/autotrain-data-company_all", "transformers", "autotrain", "co2_eq_emissions", "autotrain_compatible" ]	token-classification	false	ismail-lucifer011	null	ismail-lucifer011/autotrain-company_all-903429548	89	null	transformers	4,819	--- tags: autotrain language: en widget: - text: "I love AutoTrain 🤗" datasets: - ismail-lucifer011/autotrain-data-company_all co2_eq_emissions: 0.848790823793881 --- # Model Trained Using AutoTrain - Problem type: Entity Extraction - Model ID: 903429548 - CO2 Emissions (in grams): 0.848790823793881 ## Validation Metrics - Loss: 0.006148040760308504 - Accuracy: 0.9979930566588805 - Precision: 0.9814944904963571 - Recall: 0.9817210885036588 - F1: 0.9816077764228254 ## Usage You can use cURL to access this model: ``` $ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoTrain"}' https://api-inference.huggingface.co/models/ismail-lucifer011/autotrain-company_all-903429548 ``` Or Python API: ``` from transformers import AutoModelForTokenClassification, AutoTokenizer model = AutoModelForTokenClassification.from_pretrained("ismail-lucifer011/autotrain-company_all-903429548", use_auth_token=True) tokenizer = AutoTokenizer.from_pretrained("ismail-lucifer011/autotrain-company_all-903429548", use_auth_token=True) inputs = tokenizer("I love AutoTrain", return_tensors="pt") outputs = model(**inputs) ```
Genario/multilingual_paraphrase	40fa64fac60a68006923e19daefa6545b780888c	2022-06-27T13:01:00.000Z	[ "pytorch", "tf", "bert", "feature-extraction", "multilingual", "sentence-transformers", "sentence-similarity", "transformers", "license:apache-2.0" ]	feature-extraction	false	Genario	null	Genario/multilingual_paraphrase	89	null	sentence-transformers	4,820	--- pipeline_tag: feature-extraction language: multilingual license: apache-2.0 tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search.
nielsr/convnext-tiny-maskrcnn	4e675c60d14587de898a2803cc34dc733f9a4899	2022-06-28T10:05:05.000Z	[ "pytorch", "convnext_maskrcnn", "transformers" ]	null	false	nielsr	null	nielsr/convnext-tiny-maskrcnn	89	null	transformers	4,821	Entry not found
throwaway112358112358/DialoGPT-medium-script	890ae93de34b8c42ca3e5b21a3ef09ca410f4fd0	2022-07-22T06:23:55.000Z	[ "pytorch", "gpt2", "text-generation", "transformers", "conversational" ]	conversational	false	throwaway112358112358	null	throwaway112358112358/DialoGPT-medium-script	89	null	transformers	4,822	--- tags: - conversational --- # America DialoGPT Model
Helsinki-NLP/opus-mt-af-es	2b07ca080f32eb33642e2da486f8e5846f1bd7b7	2021-01-18T07:46:14.000Z	[ "pytorch", "marian", "text2text-generation", "af", "es", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-af-es	88	null	transformers	4,823	--- language: - af - es tags: - translation license: apache-2.0 --- ### afr-spa * source group: Afrikaans * target group: Spanish * OPUS readme: [afr-spa](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/afr-spa/README.md) * model: transformer-align * source language(s): afr * target language(s): spa * model: transformer-align * pre-processing: normalization + SentencePiece (spm32k,spm32k) * download original weights: [opus-2020-06-17.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/afr-spa/opus-2020-06-17.zip) * test set translations: [opus-2020-06-17.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/afr-spa/opus-2020-06-17.test.txt) * test set scores: [opus-2020-06-17.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/afr-spa/opus-2020-06-17.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba-test.afr.spa \| 49.9 \| 0.680 \| ### System Info: - hf_name: afr-spa - source_languages: afr - target_languages: spa - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/afr-spa/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['af', 'es'] - src_constituents: {'afr'} - tgt_constituents: {'spa'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/afr-spa/opus-2020-06-17.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/afr-spa/opus-2020-06-17.test.txt - src_alpha3: afr - tgt_alpha3: spa - short_pair: af-es - chrF2_score: 0.68 - bleu: 49.9 - brevity_penalty: 1.0 - ref_len: 2783.0 - src_name: Afrikaans - tgt_name: Spanish - train_date: 2020-06-17 - src_alpha2: af - tgt_alpha2: es - prefer_old: False - long_pair: afr-spa - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
Helsinki-NLP/opus-mt-ar-pl	623776f6e081085cf6c215c9fd91326fbf80e090	2021-01-18T07:47:35.000Z	[ "pytorch", "marian", "text2text-generation", "ar", "pl", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-ar-pl	88	null	transformers	4,824	--- language: - ar - pl tags: - translation license: apache-2.0 --- ### ara-pol * source group: Arabic * target group: Polish * OPUS readme: [ara-pol](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/ara-pol/README.md) * model: transformer * source language(s): ara arz * target language(s): pol * model: transformer * pre-processing: normalization + SentencePiece (spm32k,spm32k) * download original weights: [opus-2020-07-03.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/ara-pol/opus-2020-07-03.zip) * test set translations: [opus-2020-07-03.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/ara-pol/opus-2020-07-03.test.txt) * test set scores: [opus-2020-07-03.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/ara-pol/opus-2020-07-03.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba-test.ara.pol \| 38.0 \| 0.623 \| ### System Info: - hf_name: ara-pol - source_languages: ara - target_languages: pol - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/ara-pol/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['ar', 'pl'] - src_constituents: {'apc', 'ara', 'arq_Latn', 'arq', 'afb', 'ara_Latn', 'apc_Latn', 'arz'} - tgt_constituents: {'pol'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/ara-pol/opus-2020-07-03.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/ara-pol/opus-2020-07-03.test.txt - src_alpha3: ara - tgt_alpha3: pol - short_pair: ar-pl - chrF2_score: 0.623 - bleu: 38.0 - brevity_penalty: 0.948 - ref_len: 1171.0 - src_name: Arabic - tgt_name: Polish - train_date: 2020-07-03 - src_alpha2: ar - tgt_alpha2: pl - prefer_old: False - long_pair: ara-pol - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
Ivo/emscad-skill-extraction-token-classification	e8aea8bda132c66d6408535e2d24f75e021a3006	2021-06-15T09:37:47.000Z	[ "pytorch", "tf", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	Ivo	null	Ivo/emscad-skill-extraction-token-classification	88	null	transformers	4,825	Entry not found
dbernsohn/roberta-java	17434de4294ec97090e7a2286d5b395013b1bfe1	2021-05-20T15:54:29.000Z	[ "pytorch", "jax", "roberta", "fill-mask", "Java", "dataset:code_search_net", "arxiv:1907.11692", "transformers", "autotrain_compatible" ]	fill-mask	false	dbernsohn	null	dbernsohn/roberta-java	88	1	transformers	4,826	# roberta-java --- language: Java datasets: - code_search_net --- This is a [roberta](https://arxiv.org/pdf/1907.11692.pdf) pre-trained version on the [CodeSearchNet dataset](https://github.com/github/CodeSearchNet) for Java Mask Language Model mission. To load the model: (necessary packages: !pip install transformers sentencepiece) ```python from transformers import AutoTokenizer, AutoModelWithLMHead, pipeline tokenizer = AutoTokenizer.from_pretrained("dbernsohn/roberta-java") model = AutoModelWithLMHead.from_pretrained("dbernsohn/roberta-java") fill_mask = pipeline( "fill-mask", model=model, tokenizer=tokenizer ) ``` You can then use this model to fill masked words in a Java code. ```python code = """ String[] cars = {"Volvo", "BMW", "Ford", "Mazda"}; for (String i : cars) { System.out.<mask>(i); } """.lstrip() pred = {x["token_str"].replace("Ġ", ""): x["score"] for x in fill_mask(code)} sorted(pred.items(), key=lambda kv: kv[1], reverse=True) # [('println', 0.32571351528167725), # ('get', 0.2897663116455078), # ('remove', 0.0637081190943718), # ('exit', 0.058875661343336105), # ('print', 0.034190207719802856)] ``` The whole training process and hyperparameters are in my [GitHub repo](https://github.com/DorBernsohn/CodeLM/tree/main/CodeMLM) > Created by [Dor Bernsohn](https://www.linkedin.com/in/dor-bernsohn-70b2b1146/)
flax-community/pino-bigbird-roberta-base	81688aa15c56f034bdb694e75f526c884231de6d	2022-01-13T15:29:26.000Z	[ "pytorch", "jax", "tensorboard", "big_bird", "fill-mask", "nl", "dataset:mC4", "dataset:Dutch_news", "arxiv:2007.14062", "transformers", "autotrain_compatible" ]	fill-mask	false	flax-community	null	flax-community/pino-bigbird-roberta-base	88	1	transformers	4,827	--- language: nl datasets: - mC4 - Dutch_news --- # Pino (Dutch BigBird) base model Created by [Dat Nguyen](https://www.linkedin.com/in/dat-nguyen-49a641138/) & [Yeb Havinga](https://www.linkedin.com/in/yeb-havinga-86530825/) during the [Hugging Face community week](https://discuss.huggingface.co/t/open-to-the-community-community-week-using-jax-flax-for-nlp-cv/7104) (Not finished yet) BigBird, is a sparse-attention based transformer which extends Transformer based models, such as BERT to much longer sequences. Moreover, BigBird comes along with a theoretical understanding of the capabilities of a complete transformer that the sparse model can handle. It is a pretrained model on Dutch language using a masked language modeling (MLM) objective. It was introduced in this [paper](https://arxiv.org/abs/2007.14062) and first released in this [repository](https://github.com/google-research/bigbird). ## Model description BigBird relies on block sparse attention instead of normal attention (i.e. BERT's attention) and can handle sequences up to a length of 4096 at a much lower compute cost compared to BERT. It has achieved SOTA on various tasks involving very long sequences such as long documents summarization, question-answering with long contexts. ## How to use Here is how to use this model to get the features of a given text in PyTorch: ```python from transformers import BigBirdModel # by default its in `block_sparse` mode with num_random_blocks=3, block_size=64 model = BigBirdModel.from_pretrained("flax-community/pino-bigbird-roberta-base") # you can change `attention_type` to full attention like this: model = BigBirdModel.from_pretrained("flax-community/pino-bigbird-roberta-base", attention_type="original_full") # you can change `block_size` & `num_random_blocks` like this: model = BigBirdModel.from_pretrained("flax-community/pino-bigbird-roberta-base", block_size=16, num_random_blocks=2) ``` ## Training Data This model is pre-trained on four publicly available datasets: mC4, and scraped Dutch news from NRC en Nu.nl. It uses the the fast universal Byte-level BPE (BBPE) in contrast to the sentence piece tokenizer and vocabulary as RoBERTa (which is in turn borrowed from GPT2). ## Training Procedure The data is cleaned as follows: Remove texts containing HTML codes / javascript codes / loremipsum / policies Remove lines without end mark. Remove too short texts, words Remove too long texts, words Remove bad words ## BibTeX entry and citation info ```tex @misc{zaheer2021big, title={Big Bird: Transformers for Longer Sequences}, author={Manzil Zaheer and Guru Guruganesh and Avinava Dubey and Joshua Ainslie and Chris Alberti and Santiago Ontanon and Philip Pham and Anirudh Ravula and Qifan Wang and Li Yang and Amr Ahmed}, year={2021}, eprint={2007.14062}, archivePrefix={arXiv}, primaryClass={cs.LG} } ```
gatecitypreservation/architectural_styles	ae57ad5ef7d339c28bee56b608e4e8830284c8f3	2022-01-07T18:41:50.000Z	[ "pytorch", "tensorboard", "vit", "image-classification", "transformers", "huggingpics", "model-index" ]	image-classification	false	gatecitypreservation	null	gatecitypreservation/architectural_styles	88	1	transformers	4,828	--- tags: - image-classification - pytorch - huggingpics metrics: - accuracy model-index: - name: architectural_styles results: - task: name: Image Classification type: image-classification metrics: - name: Accuracy type: accuracy value: 0.7796609997749329 --- ### What style is that? This model can help identify five architectural styles that were prominent in the early to mid 20th century. Check back for updates including more architectural styles and more accurate predictions as this model diversifies and improves its training. Upload a photograph of a building to the File Uploader on the right. The Image Classifier will predict its architectural style using a database of over 700 images. Scroll down to read more about each style. ### Classical Revival (1895 - 1950) The Classical Revival or Neoclassical style is one of the most commonly seen across the state and the country. This style was inspired by the World's Columbian Exposition in Chicago held in 1893 which promoted a renewed interest in the classical forms. This style encompasses many different styles, including Colonial Revival, Greek Revival, Neoclassical Revival and Mediterranean Revival. Colonial Revival is most commonly used in residential dwellings, while Greek and Neoclassical Revival styles are commonly used in commercial buildings like banks, post offices, and municipal buildings. ![classical revival architecture](images/ex_classical_revival_architecture.jpg) #### Queen Anne (1880-1910) The Queen Anne style was one of a number of popular architectural styles that emerged in the United States during the Victorian Period. It ranges from high style, like the image pictured here, to more vernacular styles that exhibit the Queen Anne form without its high style architectural details. ![queen anne architecture](images/ex_queen_anne_architecture.jpg) #### Craftsman Bungalow (1900-1930) The terms “craftsman” and “bungalow” are often used interchangably, however, “craftsman” refers to the Arts and Crafts movement and is considered an architectural style, whereas “bungalow” is the form of house. Bungalows often exhibit a craftsman style. ![craftsman bungalow architecture](images/ex_craftsman_bungalow_architecture.jpg) #### Tudor Cottage (1910-1950) Tudor homes are inspired by the Medieval period and can range is size and style. In general, the Tudor style features steeply pitched roofs, often with a cat-slide roof line, predominately brick construction, sometimes accented with half-timber framing, front-facing, prominently placed brick or stone chimneys, and tall windows with rectangular or diamond-shaped panes. Front doors are typically off-center with a round arch at the top of the door or doorway. ![tudor cottage architecture](images/ex_tudor_cottage_architecture.jpg) #### Mid-Century Modern Ranch (1930-1970) The Ranch style originated in southern California in the mid-1930s. In the 1940s, the Ranch was one of the small house types financed by the Federal Housing Administration (FHA), along with Minimal Traditional and other small house styles. The Ranch house began to pick up popularity as the financial controls that encouraged small house building lifted following WWII; by the 1950s it was the most predominant residential style in the country. ![mid-century modern ranch](images/ex_mid-century_modern_ranch.jpg) This model was created with HuggingPics🤗🖼️ Image Classifier! Make your own!: [the demo on Google Colab](https://colab.research.google.com/github/nateraw/huggingpics/blob/main/HuggingPics.ipynb).
google/bert_uncased_L-2_H-512_A-8	880c57fdc84683dd9ce13a2fdbdd454abc488fb6	2021-05-19T17:29:08.000Z	[ "pytorch", "jax", "bert", "arxiv:1908.08962", "transformers", "license:apache-2.0" ]	null	false	google	null	google/bert_uncased_L-2_H-512_A-8	88	null	transformers	4,829	--- thumbnail: https://huggingface.co/front/thumbnails/google.png license: apache-2.0 --- BERT Miniatures === This is the set of 24 BERT models referenced in [Well-Read Students Learn Better: On the Importance of Pre-training Compact Models](https://arxiv.org/abs/1908.08962) (English only, uncased, trained with WordPiece masking). We have shown that the standard BERT recipe (including model architecture and training objective) is effective on a wide range of model sizes, beyond BERT-Base and BERT-Large. The smaller BERT models are intended for environments with restricted computational resources. They can be fine-tuned in the same manner as the original BERT models. However, they are most effective in the context of knowledge distillation, where the fine-tuning labels are produced by a larger and more accurate teacher. Our goal is to enable research in institutions with fewer computational resources and encourage the community to seek directions of innovation alternative to increasing model capacity. You can download the 24 BERT miniatures either from the [official BERT Github page](https://github.com/google-research/bert/), or via HuggingFace from the links below: \| \|H=128\|H=256\|H=512\|H=768\| \|---\|:---:\|:---:\|:---:\|:---:\| \| L=2 \|[2/128 (BERT-Tiny)][2_128]\|[2/256][2_256]\|[2/512][2_512]\|[2/768][2_768]\| \| L=4 \|[4/128][4_128]\|[4/256 (BERT-Mini)][4_256]\|[4/512 (BERT-Small)][4_512]\|[4/768][4_768]\| \| L=6 \|[6/128][6_128]\|[6/256][6_256]\|[6/512][6_512]\|[6/768][6_768]\| \| L=8 \|[8/128][8_128]\|[8/256][8_256]\|[8/512 (BERT-Medium)][8_512]\|[8/768][8_768]\| \| L=10 \|[10/128][10_128]\|[10/256][10_256]\|[10/512][10_512]\|[10/768][10_768]\| \| L=12 \|[12/128][12_128]\|[12/256][12_256]\|[12/512][12_512]\|[12/768 (BERT-Base)][12_768]\| Note that the BERT-Base model in this release is included for completeness only; it was re-trained under the same regime as the original model. Here are the corresponding GLUE scores on the test set: \|Model\|Score\|CoLA\|SST-2\|MRPC\|STS-B\|QQP\|MNLI-m\|MNLI-mm\|QNLI(v2)\|RTE\|WNLI\|AX\| \|---\|:---:\|:---:\|:---:\|:---:\|:---:\|:---:\|:---:\|:---:\|:---:\|:---:\|:---:\|:---:\| \|BERT-Tiny\|64.2\|0.0\|83.2\|81.1/71.1\|74.3/73.6\|62.2/83.4\|70.2\|70.3\|81.5\|57.2\|62.3\|21.0\| \|BERT-Mini\|65.8\|0.0\|85.9\|81.1/71.8\|75.4/73.3\|66.4/86.2\|74.8\|74.3\|84.1\|57.9\|62.3\|26.1\| \|BERT-Small\|71.2\|27.8\|89.7\|83.4/76.2\|78.8/77.0\|68.1/87.0\|77.6\|77.0\|86.4\|61.8\|62.3\|28.6\| \|BERT-Medium\|73.5\|38.0\|89.6\|86.6/81.6\|80.4/78.4\|69.6/87.9\|80.0\|79.1\|87.7\|62.2\|62.3\|30.5\| For each task, we selected the best fine-tuning hyperparameters from the lists below, and trained for 4 epochs: - batch sizes: 8, 16, 32, 64, 128 - learning rates: 3e-4, 1e-4, 5e-5, 3e-5 If you use these models, please cite the following paper: ``` @article{turc2019, title={Well-Read Students Learn Better: On the Importance of Pre-training Compact Models}, author={Turc, Iulia and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina}, journal={arXiv preprint arXiv:1908.08962v2 }, year={2019} } ``` [2_128]: https://huggingface.co/google/bert_uncased_L-2_H-128_A-2 [2_256]: https://huggingface.co/google/bert_uncased_L-2_H-256_A-4 [2_512]: https://huggingface.co/google/bert_uncased_L-2_H-512_A-8 [2_768]: https://huggingface.co/google/bert_uncased_L-2_H-768_A-12 [4_128]: https://huggingface.co/google/bert_uncased_L-4_H-128_A-2 [4_256]: https://huggingface.co/google/bert_uncased_L-4_H-256_A-4 [4_512]: https://huggingface.co/google/bert_uncased_L-4_H-512_A-8 [4_768]: https://huggingface.co/google/bert_uncased_L-4_H-768_A-12 [6_128]: https://huggingface.co/google/bert_uncased_L-6_H-128_A-2 [6_256]: https://huggingface.co/google/bert_uncased_L-6_H-256_A-4 [6_512]: https://huggingface.co/google/bert_uncased_L-6_H-512_A-8 [6_768]: https://huggingface.co/google/bert_uncased_L-6_H-768_A-12 [8_128]: https://huggingface.co/google/bert_uncased_L-8_H-128_A-2 [8_256]: https://huggingface.co/google/bert_uncased_L-8_H-256_A-4 [8_512]: https://huggingface.co/google/bert_uncased_L-8_H-512_A-8 [8_768]: https://huggingface.co/google/bert_uncased_L-8_H-768_A-12 [10_128]: https://huggingface.co/google/bert_uncased_L-10_H-128_A-2 [10_256]: https://huggingface.co/google/bert_uncased_L-10_H-256_A-4 [10_512]: https://huggingface.co/google/bert_uncased_L-10_H-512_A-8 [10_768]: https://huggingface.co/google/bert_uncased_L-10_H-768_A-12 [12_128]: https://huggingface.co/google/bert_uncased_L-12_H-128_A-2 [12_256]: https://huggingface.co/google/bert_uncased_L-12_H-256_A-4 [12_512]: https://huggingface.co/google/bert_uncased_L-12_H-512_A-8 [12_768]: https://huggingface.co/google/bert_uncased_L-12_H-768_A-12
google/tapas-small-finetuned-tabfact	2d1b432ff29227fabbfab71456ddec10581a163f	2021-11-29T13:07:47.000Z	[ "pytorch", "tf", "tapas", "text-classification", "en", "dataset:tab_fact", "arxiv:2010.00571", "arxiv:2004.02349", "transformers", "sequence-classification", "license:apache-2.0" ]	text-classification	false	google	null	google/tapas-small-finetuned-tabfact	88	null	transformers	4,830	--- language: en tags: - tapas - sequence-classification license: apache-2.0 datasets: - tab_fact --- # TAPAS small model fine-tuned on Tabular Fact Checking (TabFact) This model has 2 versions which can be used. The latest version, which is the default one, corresponds to the `tapas_tabfact_inter_masklm_small_reset` checkpoint of the [original Github repository](https://github.com/google-research/tapas). This model was pre-trained on MLM and an additional step which the authors call intermediate pre-training, and then fine-tuned on [TabFact](https://github.com/wenhuchen/Table-Fact-Checking). It uses relative position embeddings by default (i.e. resetting the position index at every cell of the table). The other (non-default) version which can be used is the one with absolute position embeddings: - `no_reset`, which corresponds to `tapas_tabfact_inter_masklm_small` Disclaimer: The team releasing TAPAS did not write a model card for this model so this model card has been written by the Hugging Face team and contributors. ## Model description TAPAS is a BERT-like transformers model pretrained on a large corpus of English data from Wikipedia in a self-supervised fashion. This means it was pretrained on the raw tables and associated texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was pretrained with two objectives: - Masked language modeling (MLM): taking a (flattened) table and associated context, the model randomly masks 15% of the words in the input, then runs the entire (partially masked) sequence through the model. The model then has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like GPT which internally mask the future tokens. It allows the model to learn a bidirectional representation of a table and associated text. - Intermediate pre-training: to encourage numerical reasoning on tables, the authors additionally pre-trained the model by creating a balanced dataset of millions of syntactically created training examples. Here, the model must predict (classify) whether a sentence is supported or refuted by the contents of a table. The training examples are created based on synthetic as well as counterfactual statements. This way, the model learns an inner representation of the English language used in tables and associated texts, which can then be used to extract features useful for downstream tasks such as answering questions about a table, or determining whether a sentence is entailed or refuted by the contents of a table. Fine-tuning is done by adding a classification head on top of the pre-trained model, and then jointly train this randomly initialized classification head with the base model on TabFact. ## Intended uses & limitations You can use this model for classifying whether a sentence is supported or refuted by the contents of a table. For code examples, we refer to the documentation of TAPAS on the HuggingFace website. ## Training procedure ### Preprocessing The texts are lowercased and tokenized using WordPiece and a vocabulary size of 30,000. The inputs of the model are then of the form: ``` [CLS] Sentence [SEP] Flattened table [SEP] ``` ### Fine-tuning The model was fine-tuned on 32 Cloud TPU v3 cores for 80,000 steps with maximum sequence length 512 and batch size of 512. In this setup, fine-tuning takes around 14 hours. The optimizer used is Adam with a learning rate of 2e-5, and a warmup ratio of 0.05. See the [paper](https://arxiv.org/abs/2010.00571) for more details (appendix A2). ### BibTeX entry and citation info ```bibtex @misc{herzig2020tapas, title={TAPAS: Weakly Supervised Table Parsing via Pre-training}, author={Jonathan Herzig and Paweł Krzysztof Nowak and Thomas Müller and Francesco Piccinno and Julian Martin Eisenschlos}, year={2020}, eprint={2004.02349}, archivePrefix={arXiv}, primaryClass={cs.IR} } ``` ```bibtex @misc{eisenschlos2020understanding, title={Understanding tables with intermediate pre-training}, author={Julian Martin Eisenschlos and Syrine Krichene and Thomas Müller}, year={2020}, eprint={2010.00571}, archivePrefix={arXiv}, primaryClass={cs.CL} } ``` ```bibtex @inproceedings{2019TabFactA, title={TabFact : A Large-scale Dataset for Table-based Fact Verification}, author={Wenhu Chen, Hongmin Wang, Jianshu Chen, Yunkai Zhang, Hong Wang, Shiyang Li, Xiyou Zhou and William Yang Wang}, booktitle = {International Conference on Learning Representations (ICLR)}, address = {Addis Ababa, Ethiopia}, month = {April}, year = {2020} } ```
ibombonato/swin-age-classifier	5ebf0a390d42254891441ddc0ba9a72564aaa1eb	2022-02-11T21:42:47.000Z	[ "pytorch", "tensorboard", "swin", "image-classification", "transformers", "huggingpics", "model-index" ]	image-classification	false	ibombonato	null	ibombonato/swin-age-classifier	88	null	transformers	4,831	--- tags: - image-classification - pytorch - huggingpics metrics: - accuracy model-index: - name: swin-age-classifier results: - task: name: Image Classification type: image-classification metrics: - name: Accuracy type: accuracy value: 0.8174999952316284 --- # swin-age-classifier Trained on 80 epochs - Data from: Ai Crowd - Blitz ai-blitz-xiii - Age Prediction https://www.aicrowd.com/challenges/ai-blitz-xiii/problems/age-prediction/ Notebook based on HuggingPics Autogenerated by HuggingPics🤗🖼️ Create your own image classifier for anything by running [the demo on Google Colab](https://colab.research.google.com/github/nateraw/huggingpics/blob/main/HuggingPics.ipynb). Report any issues with the demo at the [github repo](https://github.com/nateraw/huggingpics).
mrm8488/gpt2-finetuned-recipes-cooking	b4f72aae501e00b4a1ba022ef93cb5269ed9eaf7	2021-05-23T10:24:14.000Z	[ "pytorch", "jax", "gpt2", "text-generation", "en", "transformers" ]	text-generation	false	mrm8488	null	mrm8488/gpt2-finetuned-recipes-cooking	88	null	transformers	4,832	--- language: en thumbnail: widget: - text: "HuggingFace Cake:" ---
navteca/roberta-large-squad2	2434d399f7a25d47d72c7080d0e3905e6cd4bceb	2021-04-06T16:31:09.000Z	[ "pytorch", "jax", "roberta", "question-answering", "en", "dataset:squad_v2", "transformers", "license:mit", "autotrain_compatible" ]	question-answering	false	navteca	null	navteca/roberta-large-squad2	88	null	transformers	4,833	--- datasets: - squad_v2 language: en license: mit pipeline_tag: question-answering tags: - roberta - question-answering --- # Roberta large model for QA (SQuAD 2.0) This model uses [roberta-large](https://huggingface.co/roberta-large). ## Training Data The models have been trained on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset. It can be used for question answering task. ## Usage and Performance The trained model can be used like this: ```python from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline # Load model & tokenizer roberta_model = AutoModelForQuestionAnswering.from_pretrained('navteca/roberta-large-squad2') roberta_tokenizer = AutoTokenizer.from_pretrained('navteca/roberta-large-squad2') # Get predictions nlp = pipeline('question-answering', model=roberta_model, tokenizer=roberta_tokenizer) result = nlp({ 'question': 'How many people live in Berlin?', 'context': 'Berlin had a population of 3,520,031 registered inhabitants in an area of 891.82 square kilometers.' }) print(result) #{ # "answer": "3,520,031" # "end": 36, # "score": 0.96186668, # "start": 27, #} ```
prajjwal1/roberta-base-mnli	de6e05e0d9382d5344202c667368602666c8e1b2	2021-05-20T19:31:02.000Z	[ "pytorch", "jax", "roberta", "text-classification", "transformers" ]	text-classification	false	prajjwal1	null	prajjwal1/roberta-base-mnli	88	null	transformers	4,834	Roberta-base trained on MNLI. \| Task \| Accuracy \| \|---------\|----------\| \| MNLI \| 86.32 \| \| MNLI-mm \| 86.43 \| You can also check out: - `prajjwal1/roberta-base-mnli` - `prajjwal1/roberta-large-mnli` - `prajjwal1/albert-base-v2-mnli` - `prajjwal1/albert-base-v1-mnli` - `prajjwal1/albert-large-v2-mnli` [@prajjwal_1](https://twitter.com/prajjwal_1)
sagorsarker/codeswitch-hineng-pos-lince	2482a85d58bd6ac78a8d90f82bbcd600584cdd92	2021-05-19T01:06:07.000Z	[ "pytorch", "jax", "bert", "token-classification", "hi", "en", "dataset:lince", "transformers", "codeswitching", "hindi-english", "pos", "license:mit", "autotrain_compatible" ]	token-classification	false	sagorsarker	null	sagorsarker/codeswitch-hineng-pos-lince	88	null	transformers	4,835	--- language: - hi - en datasets: - lince license: mit tags: - codeswitching - hindi-english - pos --- # codeswitch-hineng-pos-lince This is a pretrained model for Part of Speech Tagging of `hindi-english` code-mixed data used from [LinCE](https://ritual.uh.edu/lince/home) This model is trained for this below repository. [https://github.com/sagorbrur/codeswitch](https://github.com/sagorbrur/codeswitch) To install codeswitch: ``` pip install codeswitch ``` ## Part-of-Speech Tagging of Hindi-English Mixed Data * Method-1 ```py from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline tokenizer = AutoTokenizer.from_pretrained("sagorsarker/codeswitch-hineng-pos-lince") model = AutoModelForTokenClassification.from_pretrained("sagorsarker/codeswitch-hineng-pos-lince") pos_model = pipeline('ner', model=model, tokenizer=tokenizer) pos_model("put any hindi english code-mixed sentence") ``` * Method-2 ```py from codeswitch.codeswitch import POS pos = POS('hin-eng') text = "" # your mixed sentence result = pos.tag(text) print(result) ```
tanmaylaud/wav2vec2-large-xlsr-hindi-marathi	269a6b04018f4d7a6740403b27ed58386bd77788	2021-04-19T18:40:07.000Z	[ "pytorch", "wav2vec2", "automatic-speech-recognition", "mr", "hi", "dataset:openslr", "dataset:interspeech_2021_asr", "transformers", "audio", "speech", "xlsr-fine-tuning-week", "hindi", "marathi", "license:apache-2.0", "model-index" ]	automatic-speech-recognition	false	tanmaylaud	null	tanmaylaud/wav2vec2-large-xlsr-hindi-marathi	88	null	transformers	4,836	--- language: [mr,hi] datasets: - openslr - interspeech_2021_asr metrics: - wer tags: - audio - automatic-speech-recognition - speech - xlsr-fine-tuning-week - hindi - marathi license: apache-2.0 model-index: - name: XLSR Wav2Vec2 Large 53 Hindi-Marathi by Tanmay Laud results: - task: name: Speech Recognition type: automatic-speech-recognition dataset: name: OpenSLR hi, OpenSLR mr type: openslr, interspeech_2021_asr metrics: - name: Test WER type: wer value: 23.736641 --- # Wav2Vec2-Large-XLSR-53-Hindi-Marathi Fine-tuned facebook/wav2vec2-large-xlsr-53 on Hindi and Marathi using the OpenSLR SLR64 datasets. When using this model, make sure that your speech input is sampled at 16kHz. ## Installation ```bash pip install git+https://github.com/huggingface/transformers.git datasets librosa torch==1.7.0 torchaudio==0.7.0 jiwer ``` ## Eval dataset: ```bash wget https://www.openslr.org/resources/103/Marathi_test.zip -P data/marathi unzip -P "K3[2?do9" data/marathi/Marathi_test.zip -d data/marathi/. tar -xzf data/marathi/Marathi_test.tar.gz -C data/marathi/. wget https://www.openslr.org/resources/103/Hindi_test.zip -P data/hindi unzip -P "w9I2{3B" data/hindi/Hindi_test.zip -d data/hindi/. tar -xzf data/hindi/Hindi_test.tar.gz -C data/hindi/. wget -O test.csv 'https://filebin.net/snrz6bt13usv8w2e/test_large.csv?t=ps3n99ho' #If download does not work, paste this link in browser: https://filebin.net/snrz6bt13usv8w2e/test_large.csv ``` ## Usage The model can be used directly (without a language model) as follows, assuming you have a dataset with Marathi text and path fields: ```python import torch import torchaudio import librosa from datasets import load_dataset from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor from datasets import load_metric, Dataset from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC wer = load_metric("wer") processor = Wav2Vec2Processor.from_pretrained('tanmaylaud/wav2vec2-large-xlsr-hindi-marathi') model = Wav2Vec2ForCTC.from_pretrained('tanmaylaud/wav2vec2-large-xlsr-hindi-marathi').to("cuda") # Preprocessing the datasets. # We need to read the audio files as arrays def speech_file_to_array_fn(batch): batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]) speech_array, sampling_rate = torchaudio.load(batch["path"]) batch["speech"] = speech_array[0].numpy() batch["sampling_rate"] = sampling_rate batch["target_text"] = batch["sentence"] batch["speech"] = librosa.resample(np.asarray(batch["speech"]), sampling_rate, 16_000) batch["sampling_rate"] = 16_000 return batch test_data= test_data.map(speech_file_to_array_fn) inputs = processor(test_data["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True) with torch.no_grad(): logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits predicted_ids = torch.argmax(logits, dim=-1) print("Prediction:", processor.batch_decode(predicted_ids)) print("Reference:", test_data["text"][:2]) ``` # Code For Evaluation on OpenSLR (Hindi + Marathi : https://filebin.net/snrz6bt13usv8w2e/test_large.csv) ```python import torchaudio import torch import librosa import numpy as np import re test = Dataset.from_csv('test.csv') chars_to_ignore_regex = '[\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\,\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\?\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\!\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\-\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\;\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\:\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\"\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\“\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\%\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\”\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\�\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\।]' # Preprocessing the datasets. # We need to read the audio files as arrays def speech_file_to_array_fn(batch): batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]) speech_array, sampling_rate = torchaudio.load(batch["path"]) batch["speech"] = speech_array[0].numpy() batch["sampling_rate"] = sampling_rate batch["target_text"] = batch["sentence"] batch["speech"] = librosa.resample(np.asarray(batch["speech"]), sampling_rate, 16_000) batch["sampling_rate"] = 16_000 return batch test= test.map(speech_file_to_array_fn) # Preprocessing the datasets. # We need to read the audio files as arrays def evaluate(batch): inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True) with torch.no_grad(): logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits pred_ids = torch.argmax(logits, dim=-1) # we do not want to group tokens when computing the metrics batch["pred_strings"] = processor.batch_decode(pred_ids) return batch test = test.map(evaluate, batched=True, batch_size=32) print("WER: {:2f}".format(100 wer.compute(predictions=test["pred_strings"], references=test["sentence"]))) ``` #### Code for Evaluation on Common Voice Hindi (Common voice does not have Marathi yet) ```python import torchaudio import torch import librosa import numpy as np import re from datasets import load_metric, load_dataset, Dataset from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC wer = load_metric("wer") processor = Wav2Vec2Processor.from_pretrained('tanmaylaud/wav2vec2-large-xlsr-hindi-marathi') model = Wav2Vec2ForCTC.from_pretrained('tanmaylaud/wav2vec2-large-xlsr-hindi-marathi').to("cuda") chars_to_ignore_regex = '[\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\,\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\?\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\.\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\!\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\-\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\;\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\:\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\"\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\“\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\%\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\”\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\�\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\।]' # Preprocessing the datasets. # We need to read the audio files as arrays def speech_file_to_array_fn(batch): batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]) speech_array, sampling_rate = torchaudio.load(batch["path"]) batch["speech"] = speech_array[0].numpy() batch["sampling_rate"] = sampling_rate batch["target_text"] = batch["sentence"] batch["speech"] = librosa.resample(np.asarray(batch["speech"]), sampling_rate, 16_000) batch["sampling_rate"] = 16_000 return batch #Run prediction on batch def evaluate(batch): inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True) with torch.no_grad(): logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits pred_ids = torch.argmax(logits, dim=-1) # we do not want to group tokens when computing the metrics batch["pred_strings"] = processor.batch_decode(pred_ids) return batch test_data = load_dataset("common_voice", "hi", split="test") test_data = test_data.map(speech_file_to_array_fn) test_data = test_data.map(evaluate, batched=True, batch_size=32) print("WER: {:2f}".format(100 * wer.compute(predictions=test_data["pred_strings"], references=test_data["sentence"]))) ``` Link to eval notebook : https://colab.research.google.com/drive/1nZRTgKfxCD9cvy90wikTHkg2il3zgcqW#scrollTo=cXWFbhb0d7DT WER : 23.736641% (OpenSLR Hindi+Marathi Test set : https://filebin.net/snrz6bt13usv8w2e/test_large.csv) WER: 44.083527% (Common Voice Hindi Test Split)
facebook/maskformer-swin-large-coco	194762ea9a98ec6d18abed4023aa346dcf1722c3	2022-04-04T16:02:13.000Z	[ "pytorch", "maskformer", "dataset:coco", "arxiv:2107.06278", "transformers", "vision", "image-segmentatiom", "license:apache-2.0" ]	null	false	facebook	null	facebook/maskformer-swin-large-coco	88	null	transformers	4,837	--- license: apache-2.0 tags: - vision - image-segmentatiom datasets: - coco --- # Mask Mask model trained on coco. It was introduced in the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) and first released in [this repository](https://github.com/facebookresearch/MaskFormer/blob/da3e60d85fdeedcb31476b5edd7d328826ce56cc/mask_former/modeling/criterion.py#L169). Disclaimer: The team releasing Mask did not write a model card for this model so this model card has been written by the Hugging Face team. ## Model description MaskFormer addresses semantic segmentation with a mask classification paradigm instead. ![model image](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/maskformer_architecture.png) ## Intended uses & limitations You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=maskformer) to look for fine-tuned versions on a task that interests you. ### How to use Here is how to use this model: ```python >>> from transformers import MaskFormerFeatureExtractor, MaskFormerForInstanceSegmentation >>> from PIL import Image >>> import requests >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> feature_extractor = MaskFormerFeatureExtractor.from_pretrained("facebook/maskformer-swin-base-ade") >>> inputs = feature_extractor(images=image, return_tensors="pt") >>> model = MaskFormerForInstanceSegmentation.from_pretrained("facebook/maskformer-swin-base-ade") >>> outputs = model(**inputs) >>> # model predicts class_queries_logits of shape `(batch_size, num_queries)` >>> # and masks_queries_logits of shape `(batch_size, num_queries, height, width)` >>> class_queries_logits = outputs.class_queries_logits >>> masks_queries_logits = outputs.masks_queries_logits >>> # you can pass them to feature_extractor for postprocessing >>> output = feature_extractor.post_process_segmentation(outputs) >>> output = feature_extractor.post_process_semantic_segmentation(outputs) >>> output = feature_extractor.post_process_panoptic_segmentation(outputs) ``` For more code examples, we refer to the [documentation](https://huggingface.co/docs/transformers/master/en/model_doc/maskformer).
cometrain/stocks-news-t5	c201c487d5821a136c7f4422b99d04d2abf68344	2022-04-14T10:08:16.000Z	[ "pytorch", "t5", "text2text-generation", "en", "dataset:financial-sentiment-analysis", "transformers", "Cometrain AutoCode", "Cometrain AlphaML", "autotrain_compatible" ]	text2text-generation	false	cometrain	null	cometrain/stocks-news-t5	88	null	transformers	4,838	--- language: - en tags: - Cometrain AutoCode - Cometrain AlphaML datasets: - financial-sentiment-analysis widget: - text: "April 14 (Reuters) - Rio Tinto (RIO.AX), one of the largest Australian mining companies, on Thursday confirmed its exit from the state mining lobby group after raising concerns that its policy on expansion of coal mines did not align with the Paris Climate Agreement." example_title: "Rio Tinto Decision (Neutral)" - text: "LONDON, April 13 (Reuters) - Crypto lender Nexo said it has teamed up with global payments company Mastercard (MA.N) to launch on Wednesday what it calls the world's first crypto-backed payment card." example_title: "New Mastercard & Nexo project (Positive)" - text: "April 14 (Reuters) - The Russian rouble weakened on Thursday, driven by expectations that Russia may relax its temporary capital control measures further, while stocks fell as the country continued what it calls 'a special military operation' in Ukraine." example_title: "Crisis in Russia (Negative)" inference: parameters: top_p: 0.9 temperature: 0.5 --- # stocks-news-t5 This model has been automatically fine-tuned and tested as part of the development of the GPT-2-based AutoML framework for accelerated and easy development of NLP enterprise solutions. Fine-tuned [T5](https://huggingface.co/t5-base) allows to analyze financial market news. Automatically trained on [Financial Sentiment Analysis(2022)](https://www.kaggle.com/datasets/sbhatti/financial-sentiment-analysis) dataset. ## Made with Cometrain AlphaML & AutoCode This model was automatically fine-tuned using the Cometrain AlphaML framework and tested with CI/CD pipeline made by Cometrain AutoCode ## Cometrain AlphaML command ```shell $ cometrain create --name stocks-news --model auto --task 'Machine learning model for finance news analysis' --output transformers ```
MiBo/SegBert	67517fe0e778a6f0a0f332725d8d784bae4ee930	2022-05-23T12:12:33.000Z	[ "pytorch", "bert", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	MiBo	null	MiBo/SegBert	88	null	transformers	4,839	Entry not found
PrimeQA/tydiqa-boolean-answer-classifier	548277c8c7a0c04468b1f99841364b731393cfe2	2022-06-28T19:52:14.000Z	[ "pytorch", "xlm-roberta", "text-classification", "arxiv:2112.07772", "arxiv:2206.08441", "transformers", "license:apache-2.0" ]	text-classification	false	PrimeQA	null	PrimeQA/tydiqa-boolean-answer-classifier	88	null	transformers	4,840	--- license: apache-2.0 --- ## Model description An answer classification model for boolean questions based on XLM-RoBERTa. The answer classifier takes as input a boolean question and a passage, and returns a label (yes, no-answer, no). The model was initialized with [xlm-roberta-large](https://huggingface.co/xlm-roberta-large) and fine-tuned on the boolean questions from [TyDiQA](https://huggingface.co/datasets/tydiqa), as well as [BoolQ-X](https://arxiv.org/abs/2112.07772#). ## Intended uses & limitations You can use the raw model for question classification. Biases associated with the pre-existing language model, xlm-roberta-large, may be present in our fine-tuned model, tydiqa-boolean-answer-classifier. ## Usage You can use this model directly in the the [PrimeQA](https://github.com/primeqa/primeqa) framework for supporting boolean questions in reading comprehension: [examples](https://github.com/primeqa/primeqa/tree/main/examples/boolqa). ### BibTeX entry and citation info ```bibtex @article{Rosenthal2021DoAT, title={Do Answers to Boolean Questions Need Explanations? Yes}, author={Sara Rosenthal and Mihaela A. Bornea and Avirup Sil and Radu Florian and Scott McCarley}, journal={ArXiv}, year={2021}, volume={abs/2112.07772} } ``` ```bibtex @misc{https://doi.org/10.48550/arxiv.2206.08441, author = {McCarley, Scott and Bornea, Mihaela and Rosenthal, Sara and Ferritto, Anthony and Sultan, Md Arafat and Sil, Avirup and Florian, Radu}, title = {GAAMA 2.0: An Integrated System that Answers Boolean and Extractive Questions}, journal = {CoRR}, publisher = {arXiv}, year = {2022}, url = {https://arxiv.org/abs/2206.08441}, } ```
ml6team/keyphrase-extraction-kbir-openkp	f17ec53f7000ea5f7d760ef6bb49efe5e2af49ca	2022-06-17T06:45:06.000Z	[ "pytorch", "roberta", "token-classification", "en", "dataset:midas/openkp", "arxiv:2112.08547", "arxiv:1911.02671", "transformers", "keyphrase-extraction", "license:mit", "model-index", "autotrain_compatible" ]	token-classification	false	ml6team	null	ml6team/keyphrase-extraction-kbir-openkp	88	null	transformers	4,841	--- language: en license: mit tags: - keyphrase-extraction datasets: - midas/openkp metrics: - seqeval widget: - text: "Keyphrase extraction is a technique in text analysis where you extract the important keyphrases from a document. Thanks to these keyphrases humans can understand the content of a text very quickly and easily without reading it completely. Keyphrase extraction was first done primarily by human annotators, who read the text in detail and then wrote down the most important keyphrases. The disadvantage is that if you work with a lot of documents, this process can take a lot of time. Here is where Artificial Intelligence comes in. Currently, classical machine learning methods, that use statistical and linguistic features, are widely used for the extraction process. Now with deep learning, it is possible to capture the semantic meaning of a text even better than these classical methods. Classical methods look at the frequency, occurrence and order of words in the text, whereas these neural approaches can capture long-term semantic dependencies and context of words in a text." example_title: "Example 1" - text: "FoodEx is the largest trade exhibition for food and drinks in Asia, with about 70,000 visitors checking out the products presented by hundreds of participating companies. I was lucky to enter as press; otherwise, visitors must be affiliated with the food industry— and pay ¥5,000 — to enter. The FoodEx menu is global, including everything from cherry beer from Germany and premium Mexican tequila to top-class French and Chinese dumplings. The event was a rare chance to try out both well-known and exotic foods and even see professionals making them. In addition to booths offering traditional Japanese favorites such as udon and maguro sashimi, there were plenty of innovative twists, such as dorayaki , a sweet snack made of two pancakes and a red-bean filling, that came in coffee and tomato flavors. While I was there I was lucky to catch the World Sushi Cup Japan 2013, where top chefs from around the world were competing … and presenting a wide range of styles that you would not normally see in Japan, like the flower makizushi above." example_title: "Example 2" model-index: - name: ml6team/keyphrase-extraction-kbir-openkp results: - task: type: keyphrase-extraction name: Keyphrase Extraction dataset: type: midas/openkp name: openkp metrics: - type: F1 (Seqeval) value: 0.000 name: F1 (Seqeval) - type: F1@M value: 0.387 name: F1@M --- # 🔑 Keyphrase Extraction Model: KBIR-OpenKP Keyphrase extraction is a technique in text analysis where you extract the important keyphrases from a document. Thanks to these keyphrases humans can understand the content of a text very quickly and easily without reading it completely. Keyphrase extraction was first done primarily by human annotators, who read the text in detail and then wrote down the most important keyphrases. The disadvantage is that if you work with a lot of documents, this process can take a lot of time ⏳. Here is where Artificial Intelligence 🤖 comes in. Currently, classical machine learning methods, that use statistical and linguistic features, are widely used for the extraction process. Now with deep learning, it is possible to capture the semantic meaning of a text even better than these classical methods. Classical methods look at the frequency, occurrence and order of words in the text, whereas these neural approaches can capture long-term semantic dependencies and context of words in a text. ## 📓 Model Description This model uses [KBIR](https://huggingface.co/bloomberg/KBIR) as its base model and fine-tunes it on the [OpenKP dataset](https://huggingface.co/datasets/midas/openkp). KBIR or Keyphrase Boundary Infilling with Replacement is a pre-trained model which utilizes a multi-task learning setup for optimizing a combined loss of Masked Language Modeling (MLM), Keyphrase Boundary Infilling (KBI) and Keyphrase Replacement Classification (KRC). You can find more information about the architecture in this [paper](https://arxiv.org/abs/2112.08547). Keyphrase extraction models are transformer models fine-tuned as a token classification problem where each word in the document is classified as being part of a keyphrase or not. \| Label \| Description \| \| ----- \| ------------------------------- \| \| B-KEY \| At the beginning of a keyphrase \| \| I-KEY \| Inside a keyphrase \| \| O \| Outside a keyphrase \| ## ✋ Intended Uses & Limitations ### 🛑 Limitations * Limited amount of predicted keyphrases. * Only works for English documents. * For a custom model, please consult the [training notebook]() for more information. ### ❓ How To Use ```python from transformers import ( TokenClassificationPipeline, AutoModelForTokenClassification, AutoTokenizer, ) from transformers.pipelines import AggregationStrategy import numpy as np # Define keyphrase extraction pipeline class KeyphraseExtractionPipeline(TokenClassificationPipeline): def __init__(self, model, args, kwargs): super().__init__( model=AutoModelForTokenClassification.from_pretrained(model), tokenizer=AutoTokenizer.from_pretrained(model), args, **kwargs ) def postprocess(self, model_outputs): results = super().postprocess( model_outputs=model_outputs, aggregation_strategy=AggregationStrategy.SIMPLE, ) return np.unique([result.get("word").strip() for result in results]) ``` ```python # Load pipeline model_name = "ml6team/keyphrase-extraction-kbir-openkp" extractor = KeyphraseExtractionPipeline(model=model_name) ``` ```python # Inference text = """ Keyphrase extraction is a technique in text analysis where you extract the important keyphrases from a document. Thanks to these keyphrases humans can understand the content of a text very quickly and easily without reading it completely. Keyphrase extraction was first done primarily by human annotators, who read the text in detail and then wrote down the most important keyphrases. The disadvantage is that if you work with a lot of documents, this process can take a lot of time. Here is where Artificial Intelligence comes in. Currently, classical machine learning methods, that use statistical and linguistic features, are widely used for the extraction process. Now with deep learning, it is possible to capture the semantic meaning of a text even better than these classical methods. Classical methods look at the frequency, occurrence and order of words in the text, whereas these neural approaches can capture long-term semantic dependencies and context of words in a text. """.replace(" ", " ") keyphrases = extractor(text) print(keyphrases) ``` ``` # Output ['keyphrase extraction' 'text analysis'] ``` ## 📚 Training Dataset [OpenKP](https://github.com/microsoft/OpenKP) is a large-scale, open-domain keyphrase extraction dataset with 148,124 real-world web documents along with 1-3 most relevant human-annotated keyphrases. You can find more information in the [paper](https://arxiv.org/abs/1911.02671). ## 👷‍♂️ Training Procedure For more in detail information, you can take a look at the [training notebook](). ### Training Parameters \| Parameter \| Value \| \| --------- \| ------\| \| Learning Rate \| 1e-4 \| \| Epochs \| 50 \| \| Early Stopping Patience \| 3 \| ### Preprocessing The documents in the dataset are already preprocessed into list of words with the corresponding labels. The only thing that must be done is tokenization and the realignment of the labels so that they correspond with the right subword tokens. ```python from datasets import load_dataset from transformers import AutoTokenizer # Labels label_list = ["B", "I", "O"] lbl2idx = {"B": 0, "I": 1, "O": 2} idx2label = {0: "B", 1: "I", 2: "O"} # Tokenizer tokenizer = AutoTokenizer.from_pretrained("bloomberg/KBIR") max_length = 512 # Dataset parameters dataset_full_name = "midas/openkp" dataset_subset = "raw" dataset_document_column = "document" dataset_biotags_column = "doc_bio_tags" def preprocess_fuction(all_samples_per_split): tokenized_samples = tokenizer.batch_encode_plus( all_samples_per_split[dataset_document_column], padding="max_length", truncation=True, is_split_into_words=True, max_length=max_length, ) total_adjusted_labels = [] for k in range(0, len(tokenized_samples["input_ids"])): prev_wid = -1 word_ids_list = tokenized_samples.word_ids(batch_index=k) existing_label_ids = all_samples_per_split[dataset_biotags_column][k] i = -1 adjusted_label_ids = [] for wid in word_ids_list: if wid is None: adjusted_label_ids.append(lbl2idx["O"]) elif wid != prev_wid: i = i + 1 adjusted_label_ids.append(lbl2idx[existing_label_ids[i]]) prev_wid = wid else: adjusted_label_ids.append( lbl2idx[ f"{'I' if existing_label_ids[i] == 'B' else existing_label_ids[i]}" ] ) total_adjusted_labels.append(adjusted_label_ids) tokenized_samples["labels"] = total_adjusted_labels return tokenized_samples # Load dataset dataset = load_dataset(dataset_full_name, dataset_subset) # Preprocess dataset tokenized_dataset = dataset.map(preprocess_fuction, batched=True) ``` ### Postprocessing (Without Pipeline Function) If you do not use the pipeline function, you must filter out the B and I labeled tokens. Each B and I will then be merged into a keyphrase. Finally, you need to strip the keyphrases to make sure all unnecessary spaces have been removed. ```python # Define post_process functions def concat_tokens_by_tag(keyphrases): keyphrase_tokens = [] for id, label in keyphrases: if label == "B": keyphrase_tokens.append([id]) elif label == "I": if len(keyphrase_tokens) > 0: keyphrase_tokens[len(keyphrase_tokens) - 1].append(id) return keyphrase_tokens def extract_keyphrases(example, predictions, tokenizer, index=0): keyphrases_list = [ (id, idx2label[label]) for id, label in zip( np.array(example["input_ids"]).squeeze().tolist(), predictions[index] ) if idx2label[label] in ["B", "I"] ] processed_keyphrases = concat_tokens_by_tag(keyphrases_list) extracted_kps = tokenizer.batch_decode( processed_keyphrases, skip_special_tokens=True, clean_up_tokenization_spaces=True, ) return np.unique([kp.strip() for kp in extracted_kps]) ``` ## 📝 Evaluation Results Traditional evaluation methods are the precision, recall and F1-score @k,m where k is the number that stands for the first k predicted keyphrases and m for the average amount of predicted keyphrases. The model achieves the following results on the OpenKP test set: \| Dataset \| P@5 \| R@5 \| F1@5 \| P@10 \| R@10 \| F1@10 \| P@M \| R@M \| F1@M \| \|:-----------------:\|:----:\|:----:\|:----:\|:----:\|:----:\|:-----:\|:----:\|:----:\|:----:\| \| OpenKP Test Set \| 0.13 \| 0.38 \| 0.19 \| 0.07 \| 0.38 \| 0.11 \| 0.45 \| 0.38 \| 0.39 \| For more information on the evaluation process, you can take a look at the keyphrase extraction evaluation notebook. ## 🚨 Issues Please feel free to start discussions in the Community Tab.
hakurei/litv2-6B-rev3	b3149de232b8f398645015eaef23c2dd886f244b	2022-06-17T04:54:20.000Z	[ "pytorch", "gptj", "text-generation", "transformers" ]	text-generation	false	hakurei	null	hakurei/litv2-6B-rev3	88	null	transformers	4,842	https://wandb.ai/haruu/mesh-transformer-jax/runs/1iae931p?workspace=user-haruu
davidcechak/DNADebertaK8	e34c774d03fc1c65d877d21134f86a5c58ac35e9	2022-07-05T22:54:48.000Z	[ "pytorch", "tensorboard", "deberta", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	davidcechak	null	davidcechak/DNADebertaK8	88	null	transformers	4,843	Entry not found
CAMeL-Lab/bert-base-arabic-camelbert-ca-sentiment	a5482ef3f4cf508d383ef4a3cc7ce40cfa12722b	2021-10-17T11:15:12.000Z	[ "pytorch", "tf", "bert", "text-classification", "ar", "arxiv:2103.06678", "transformers", "license:apache-2.0" ]	text-classification	false	CAMeL-Lab	null	CAMeL-Lab/bert-base-arabic-camelbert-ca-sentiment	87	2	transformers	4,844	--- language: - ar license: apache-2.0 widget: - text: "أنا بخير" --- # CAMeLBERT-CA SA Model ## Model description CAMeLBERT-CA SA Model is a Sentiment Analysis (SA) model that was built by fine-tuning the [CAMeLBERT Classical Arabic (CA)](https://huggingface.co/CAMeL-Lab/bert-base-arabic-camelbert-ca/) model. For the fine-tuning, we used the [ASTD](https://aclanthology.org/D15-1299.pdf), [ArSAS](http://lrec-conf.org/workshops/lrec2018/W30/pdf/22_W30.pdf), and [SemEval](https://aclanthology.org/S17-2088.pdf) datasets. Our fine-tuning procedure and the hyperparameters we used can be found in our paper "[The Interplay of Variant, Size, and Task Type in Arabic Pre-trained Language Models](https://arxiv.org/abs/2103.06678)." Our fine-tuning code can be found [here](https://github.com/CAMeL-Lab/CAMeLBERT). ## Intended uses You can use the CAMeLBERT-CA SA model directly as part of our [CAMeL Tools](https://github.com/CAMeL-Lab/camel_tools) SA component (recommended) or as part of the transformers pipeline. #### How to use To use the model with the [CAMeL Tools](https://github.com/CAMeL-Lab/camel_tools) SA component: ```python >>> from camel_tools.sentiment import SentimentAnalyzer >>> sa = SentimentAnalyzer("CAMeL-Lab/bert-base-arabic-camelbert-ca-sentiment") >>> sentences = ['أنا بخير', 'أنا لست بخير'] >>> sa.predict(sentences) >>> ['positive', 'negative'] ``` You can also use the SA model directly with a transformers pipeline: ```python >>> from transformers import pipeline e >>> sa = pipeline('text-classification', model='CAMeL-Lab/bert-base-arabic-camelbert-ca-sentiment') >>> sentences = ['أنا بخير', 'أنا لست بخير'] >>> sa(sentences) [{'label': 'positive', 'score': 0.9616648554801941}, {'label': 'negative', 'score': 0.9779177904129028}] ``` Note: to download our models, you would need `transformers>=3.5.0`. Otherwise, you could download the models manually. ## Citation ```bibtex @inproceedings{inoue-etal-2021-interplay, title = "The Interplay of Variant, Size, and Task Type in {A}rabic Pre-trained Language Models", author = "Inoue, Go and Alhafni, Bashar and Baimukan, Nurpeiis and Bouamor, Houda and Habash, Nizar", booktitle = "Proceedings of the Sixth Arabic Natural Language Processing Workshop", month = apr, year = "2021", address = "Kyiv, Ukraine (Online)", publisher = "Association for Computational Linguistics", abstract = "In this paper, we explore the effects of language variants, data sizes, and fine-tuning task types in Arabic pre-trained language models. To do so, we build three pre-trained language models across three variants of Arabic: Modern Standard Arabic (MSA), dialectal Arabic, and classical Arabic, in addition to a fourth language model which is pre-trained on a mix of the three. We also examine the importance of pre-training data size by building additional models that are pre-trained on a scaled-down set of the MSA variant. We compare our different models to each other, as well as to eight publicly available models by fine-tuning them on five NLP tasks spanning 12 datasets. Our results suggest that the variant proximity of pre-training data to fine-tuning data is more important than the pre-training data size. We exploit this insight in defining an optimized system selection model for the studied tasks.", } ```
Helsinki-NLP/opus-mt-fr-pl	e7d23016ef0cf42510b2a7701e36b8b08b96399c	2021-09-09T21:56:11.000Z	[ "pytorch", "marian", "text2text-generation", "fr", "pl", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-fr-pl	87	null	transformers	4,845	--- tags: - translation license: apache-2.0 --- ### opus-mt-fr-pl * source languages: fr * target languages: pl * OPUS readme: [fr-pl](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/fr-pl/README.md) * dataset: opus * model: transformer-align * pre-processing: normalization + SentencePiece * download original weights: [opus-2020-01-16.zip](https://object.pouta.csc.fi/OPUS-MT-models/fr-pl/opus-2020-01-16.zip) * test set translations: [opus-2020-01-16.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/fr-pl/opus-2020-01-16.test.txt) * test set scores: [opus-2020-01-16.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/fr-pl/opus-2020-01-16.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba.fr.pl \| 40.7 \| 0.625 \|
Helsinki-NLP/opus-mt-ja-pt	654bbe699ed24805d8f5155246b278b3fa65acb9	2020-08-21T14:42:47.000Z	[ "pytorch", "marian", "text2text-generation", "ja", "pt", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-ja-pt	87	null	transformers	4,846	--- language: - ja - pt tags: - translation license: apache-2.0 --- ### jpn-por * source group: Japanese * target group: Portuguese * OPUS readme: [jpn-por](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/jpn-por/README.md) * model: transformer-align * source language(s): jpn jpn_Hani jpn_Hira jpn_Kana jpn_Latn jpn_Yiii * target language(s): por por_Hira * model: transformer-align * pre-processing: normalization + SentencePiece (spm32k,spm32k) * a sentence initial language token is required in the form of `>>id<<` (id = valid target language ID) * download original weights: [opus-2020-06-17.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-por/opus-2020-06-17.zip) * test set translations: [opus-2020-06-17.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-por/opus-2020-06-17.test.txt) * test set scores: [opus-2020-06-17.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-por/opus-2020-06-17.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba-test.jpn.por \| 22.2 \| 0.444 \| ### System Info: - hf_name: jpn-por - source_languages: jpn - target_languages: por - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/jpn-por/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['ja', 'pt'] - src_constituents: {'jpn_Hang', 'jpn', 'jpn_Yiii', 'jpn_Kana', 'jpn_Hani', 'jpn_Bopo', 'jpn_Latn', 'jpn_Hira'} - tgt_constituents: {'por'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-por/opus-2020-06-17.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-por/opus-2020-06-17.test.txt - src_alpha3: jpn - tgt_alpha3: por - short_pair: ja-pt - chrF2_score: 0.444 - bleu: 22.2 - brevity_penalty: 0.922 - ref_len: 15570.0 - src_name: Japanese - tgt_name: Portuguese - train_date: 2020-06-17 - src_alpha2: ja - tgt_alpha2: pt - prefer_old: False - long_pair: jpn-por - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
NLP4H/ms_bert	7883a03983cc71f12af68e838fad140865cbf97f	2021-05-18T21:46:48.000Z	[ "pytorch", "jax", "bert", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	NLP4H	null	NLP4H/ms_bert	87	null	transformers	4,847	# MS-BERT ## Introduction This repository provides codes and models of MS-BERT. MS-BERT was pre-trained on notes from neurological examination for Multiple Sclerosis (MS) patients at St. Michael's Hospital in Toronto, Canada. ## Data The dataset contained approximately 75,000 clinical notes, for about 5000 patients, totaling to over 35.7 million words. These notes were collected from patients who visited St. Michael's Hospital MS Clinic between 2015 to 2019. The notes contained a variety of information pertaining to a neurological exam. For example, a note can contain information on the patient's condition, their progress over time and diagnosis. The gender split within the dataset was observed to be 72% female and 28% male ([which reflects the natural discrepancy seen in MS][1]). Further sections will describe how MS-BERT was pre trained through the use of these clinically relevant and rich neurological notes. ## Data pre-processing The data was pre-processed to remove any identifying information. This includes information on: patient names, doctor names, hospital names, patient identification numbers, phone numbers, addresses, and time. In order to de-identify the information, we used a curated database that contained patient and doctor information. This curated database was paired with regular expressions to find and remove any identifying pieces of information. Each of these identifiers were replaced with a specific token. These tokens were chosen based on three criteria: (1) they belong to the current BERT vocab, (2), they have relatively the same semantic meaning as the word they are replacing, and (3), the token is not found in the original unprocessed dataset. The replacements that met the criteria above were as follows: Female first names -> Lucie Male first names -> Ezekiel Last/family names -> Salamanca. Dates -> 2010s Patient IDs -> 999 Phone numbers -> 1718 Addresses -> Silesia Time -> 1610 Locations/Hospital/Clinic names -> Troy ## Pre-training The starting point for our model is the already pre-trained and fine-tuned BLUE-BERT base. We further pre-train it using the masked language modelling task from the huggingface transformers [library](https://github.com/huggingface). The hyperparameters can be found in the config file in this repository or [here](https://s3.amazonaws.com/models.huggingface.co/bert/NLP4H/ms_bert/config.json) ## Acknowledgements We would like to thank the researchers and staff at the Data Science and Advanced Analytics (DSAA) department, St. Michael’s Hospital, for providing consistent support and guidance throughout this project. We would also like to thank Dr. Marzyeh Ghassemi, Taylor Killan, Nathan Ng and Haoran Zhang for providing us the opportunity to work on this exciting project. ## Disclaimer MS-BERT shows the results of research conducted at the Data Science and Advanced Analytics (DSAA) department, St. Michael’s Hospital. The results produced by MS-BERT are not intended for direct diagnostic use or medical decision-making without review and oversight by a clinical professional. Individuals should not make decisions about their health solely on the basis of the results produced by MS-BERT. St. Michael’s Hospital does not independently verify the validity or utility of the results produced by MS-BERT. If you have questions about the results produced by MS-BERT please consult a healthcare professional. If you would like more information about the research conducted at DSAA please contact [Zhen Yang](mailto:[email protected]). If you would like more information on neurological examination notes please contact [Dr. Tony Antoniou](mailto:[email protected]) or [Dr. Jiwon Oh](mailto:[email protected]) from the MS clinic at St. Michael's Hospital. [1]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3707353/
NbAiLab/nb-bert-base-ner	b925de07fe9e9dc55c20cd72364b791aef42e2d0	2022-06-01T10:49:24.000Z	[ "pytorch", "bert", "token-classification", "no", "dataset:norne", "transformers", "norwegian", "ner", "license:cc-by-4.0", "autotrain_compatible" ]	token-classification	false	NbAiLab	null	NbAiLab/nb-bert-base-ner	87	null	transformers	4,848	--- language: no license: cc-by-4.0 tags: - norwegian - bert - ner thumbnail: nblogo_3.png pipeline_tag: token-classification datasets: - norne inference: parameters: aggregation_strategy: "first" widget: - text: Trond Giske har bekreftet på spørsmål fra Adresseavisen at Hansen leide et rom i hans leilighet i Trondheim. --- Release 1.0 (November 17, 2021) # nb-bert-base-ner ## Description NB-Bert base model fine-tuned on the Named Entity Recognition task using the [NorNE dataset](https://huggingface.co/datasets/NbAiLab/norne). ## Usage ```python from transformers import AutoTokenizer, AutoModelForTokenClassification from transformers import pipeline tokenizer = AutoTokenizer.from_pretrained("NbAiLab/nb-bert-base-ner") model = AutoModelForTokenClassification.from_pretrained("NbAiLab/nb-bert-base-ner") nlp = pipeline("ner", model=model, tokenizer=tokenizer) example = "Jeg heter Kjell og bor i Oslo." ner_results = nlp(example) print(ner_results) ```
Saz/DialoGPT-small-saz	10167fd59643d1389903b0f470c64bae721de24c	2021-10-08T06:08:56.000Z	[ "pytorch", "gpt2", "text-generation", "transformers", "conversational" ]	conversational	false	Saz	null	Saz/DialoGPT-small-saz	87	null	transformers	4,849	--- tags: - conversational --- # Saz DialoGPT Model
UBC-NLP/AraT5-tweet-base	8926762f3fdf3661c8b632f368e1a5f5b4c5a1ab	2022-05-26T18:26:55.000Z	[ "pytorch", "tf", "t5", "ar", "transformers", "Arabic T5", "MSA", "Twitter", "Arabic Dialect", "Arabic Machine Translation", "Arabic Text Summarization", "Arabic News Title and Question Generation", "Arabic Paraphrasing and Transliteration", "Arabic Code-Switched Translation" ]	null	false	UBC-NLP	null	UBC-NLP/AraT5-tweet-base	87	1	transformers	4,850	--- language: - ar tags: - Arabic T5 - MSA - Twitter - Arabic Dialect - Arabic Machine Translation - Arabic Text Summarization - Arabic News Title and Question Generation - Arabic Paraphrasing and Transliteration - Arabic Code-Switched Translation --- # AraT5-base # AraT5: Text-to-Text Transformers for Arabic Language Generation <img src="https://huggingface.co/UBC-NLP/AraT5-base/resolve/main/AraT5_CR_new.png" alt="AraT5" width="45%" height="35%" align="right"/> This is the repository accompanying our paper [AraT5: Text-to-Text Transformers for Arabic Language Understanding and Generation](https://aclanthology.org/2022.acl-long.47/). In this is the repository we Introduce AraT5<sub>MSA</sub>, AraT5<sub>Tweet</sub>, and AraT5: three powerful Arabic-specific text-to-text Transformer based models; --- # How to use AraT5 models Below is an example for fine-tuning AraT5-base for News Title Generation on the Aranews dataset ``` bash !python run_trainier_seq2seq_huggingface.py \ --learning_rate 5e-5 \ --max_target_length 128 --max_source_length 128 \ --per_device_train_batch_size 8 --per_device_eval_batch_size 8 \ --model_name_or_path "UBC-NLP/AraT5-base" \ --output_dir "/content/AraT5_FT_title_generation" --overwrite_output_dir \ --num_train_epochs 3 \ --train_file "/content/ARGEn_title_genration_sample_train.tsv" \ --validation_file "/content/ARGEn_title_genration_sample_valid.tsv" \ --task "title_generation" --text_column "document" --summary_column "title" \ --load_best_model_at_end --metric_for_best_model "eval_bleu" --greater_is_better True --evaluation_strategy epoch --logging_strategy epoch --predict_with_generate\ --do_train --do_eval ``` For more details about the fine-tuning example, please read this notebook [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://github.com/UBC-NLP/araT5/blob/main/examples/Fine_tuning_AraT5.ipynb) In addition, we release the fine-tuned checkpoint of the News Title Generation (NGT) which is described in the paper. The model available at Huggingface ([UBC-NLP/AraT5-base-title-generation](https://huggingface.co/UBC-NLP/AraT5-base-title-generation)). For more details, please visit our own [GitHub](https://github.com/UBC-NLP/araT5). # AraT5 Models Checkpoints AraT5 Pytorch and TensorFlow checkpoints are available on the Huggingface website for direct download and use ```exclusively for research```. ```For commercial use, please contact the authors via email @ (muhammad.mageed[at]ubc[dot]ca).``` \| Model \| Link \| \|---------\|:------------------:\| \| AraT5-base \| [https://huggingface.co/UBC-NLP/AraT5-base](https://huggingface.co/UBC-NLP/AraT5-base) \| \| AraT5-msa-base \| [https://huggingface.co/UBC-NLP/AraT5-msa-base](https://huggingface.co/UBC-NLP/AraT5-msa-base) \| \| AraT5-tweet-base \| [https://huggingface.co/UBC-NLP/AraT5-tweet-base](https://huggingface.co/UBC-NLP/AraT5-tweet-base) \| \| AraT5-msa-small \| [https://huggingface.co/UBC-NLP/AraT5-msa-small](https://huggingface.co/UBC-NLP/AraT5-msa-small) \| \| AraT5-tweet-small\| [https://huggingface.co/UBC-NLP/AraT5-tweet-small](https://huggingface.co/UBC-NLP/AraT5-tweet-small) \| # BibTex If you use our models (Arat5-base, Arat5-msa-base, Arat5-tweet-base, Arat5-msa-small, or Arat5-tweet-small ) for your scientific publication, or if you find the resources in this repository useful, please cite our paper as follows (to be updated): ```bibtex @inproceedings{nagoudi-etal-2022-arat5, title = "{A}ra{T}5: Text-to-Text Transformers for {A}rabic Language Generation", author = "Nagoudi, El Moatez Billah and Elmadany, AbdelRahim and Abdul-Mageed, Muhammad", booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = may, year = "2022", address = "Dublin, Ireland", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2022.acl-long.47", pages = "628--647", abstract = "Transfer learning with a unified Transformer framework (T5) that converts all language problems into a text-to-text format was recently proposed as a simple and effective transfer learning approach. Although a multilingual version of the T5 model (mT5) was also introduced, it is not clear how well it can fare on non-English tasks involving diverse data. To investigate this question, we apply mT5 on a language with a wide variety of dialects{--}Arabic. For evaluation, we introduce a novel benchmark for ARabic language GENeration (ARGEN), covering seven important tasks. For model comparison, we pre-train three powerful Arabic T5-style models and evaluate them on ARGEN. Although pre-trained with {\textasciitilde}49 less data, our new models perform significantly better than mT5 on all ARGEN tasks (in 52 out of 59 test sets) and set several new SOTAs. Our models also establish new SOTA on the recently-proposed, large Arabic language understanding evaluation benchmark ARLUE (Abdul-Mageed et al., 2021). Our new models are publicly available. We also link to ARGEN datasets through our repository: https://github.com/UBC-NLP/araT5.", } ``` ## Acknowledgments We gratefully acknowledge support from the Natural Sciences and Engineering Research Council of Canada, the Social Sciences and Humanities Research Council of Canada, Canadian Foundation for Innovation, [ComputeCanada](www.computecanada.ca) and [UBC ARC-Sockeye](https://doi.org/10.14288/SOCKEYE). We also thank the [Google TensorFlow Research Cloud (TFRC)](https://www.tensorflow.org/tfrc) program for providing us with free TPU access.
facebook/detr-resnet-50-dc5	e3ea9b76022e0df3be936441d20ee5470886f557	2022-06-27T08:36:56.000Z	[ "pytorch", "detr", "object-detection", "dataset:coco", "arxiv:2005.12872", "transformers", "vision", "license:apache-2.0" ]	object-detection	false	facebook	null	facebook/detr-resnet-50-dc5	87	1	transformers	4,851	--- license: apache-2.0 tags: - object-detection - vision datasets: - coco widget: - src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/savanna.jpg example_title: Savanna - src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/football-match.jpg example_title: Football Match - src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/airport.jpg example_title: Airport --- # DETR (End-to-End Object Detection) model with ResNet-50 backbone (dilated C5 stage) DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 object detection (118k annotated images). It was introduced in the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Carion et al. and first released in [this repository](https://github.com/facebookresearch/detr). Disclaimer: The team releasing DETR did not write a model card for this model so this model card has been written by the Hugging Face team. ## Model description The DETR model is an encoder-decoder transformer with a convolutional backbone. Two heads are added on top of the decoder outputs in order to perform object detection: a linear layer for the class labels and a MLP (multi-layer perceptron) for the bounding boxes. The model uses so-called object queries to detect objects in an image. Each object query looks for a particular object in the image. For COCO, the number of object queries is set to 100. The model is trained using a "bipartite matching loss": one compares the predicted classes + bounding boxes of each of the N = 100 object queries to the ground truth annotations, padded up to the same length N (so if an image only contains 4 objects, 96 annotations will just have a "no object" as class and "no bounding box" as bounding box). The Hungarian matching algorithm is used to create an optimal one-to-one mapping between each of the N queries and each of the N annotations. Next, standard cross-entropy (for the classes) and a linear combination of the L1 and generalized IoU loss (for the bounding boxes) are used to optimize the parameters of the model. ## Intended uses & limitations You can use the raw model for object detection. See the [model hub](https://huggingface.co/models?search=facebook/detr) to look for all available DETR models. ### How to use Here is how to use this model: ```python from transformers import DetrFeatureExtractor, DetrForObjectDetection from PIL import Image import requests url = 'http://images.cocodataset.org/val2017/000000039769.jpg' image = Image.open(requests.get(url, stream=True).raw) feature_extractor = DetrFeatureExtractor.from_pretrained('facebook/detr-resnet-50-dc5') model = DetrForObjectDetection.from_pretrained('facebook/detr-resnet-50-dc5') inputs = feature_extractor(images=image, return_tensors="pt") outputs = model(inputs) # model predicts bounding boxes and corresponding COCO classes logits = outputs.logits bboxes = outputs.pred_boxes ``` Currently, both the feature extractor and model support PyTorch. ## Training data The DETR model was trained on [COCO 2017 object detection](https://cocodataset.org/#download), a dataset consisting of 118k/5k annotated images for training/validation respectively. ## Training procedure ### Preprocessing The exact details of preprocessing of images during training/validation can be found [here](https://github.com/google-research/vision_transformer/blob/master/vit_jax/input_pipeline.py). Images are resized/rescaled such that the shortest side is at least 800 pixels and the largest side at most 1333 pixels, and normalized across the RGB channels with the ImageNet mean (0.485, 0.456, 0.406) and standard deviation (0.229, 0.224, 0.225). ### Training The model was trained for 300 epochs on 16 V100 GPUs. This takes 3 days, with 4 images per GPU (hence a total batch size of 64). ## Evaluation results This model achieves an AP (average precision) of 43.3** on COCO 2017 validation. For more details regarding evaluation results, we refer to table 1 of the original paper. ### BibTeX entry and citation info ```bibtex @article{DBLP:journals/corr/abs-2005-12872, author = {Nicolas Carion and Francisco Massa and Gabriel Synnaeve and Nicolas Usunier and Alexander Kirillov and Sergey Zagoruyko}, title = {End-to-End Object Detection with Transformers}, journal = {CoRR}, volume = {abs/2005.12872}, year = {2020}, url = {https://arxiv.org/abs/2005.12872}, archivePrefix = {arXiv}, eprint = {2005.12872}, timestamp = {Thu, 28 May 2020 17:38:09 +0200}, biburl = {https://dblp.org/rec/journals/corr/abs-2005-12872.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} } ```
franklu/pubmed_bert_squadv2	3abeb3561a41c7960a4b24b4a9629d06d9f2254b	2021-07-09T05:25:26.000Z	[ "pytorch", "bert", "question-answering", "transformers", "autotrain_compatible" ]	question-answering	false	franklu	null	franklu/pubmed_bert_squadv2	87	null	transformers	4,852	[`microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext`](https://huggingface.co/microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext) fine-tuned on [`SQuAD V2`](https://rajpurkar.github.io/SQuAD-explorer/) using [`run_qa.py`](https://github.com/huggingface/transformers/blob/master/examples/pytorch/question-answering/run_qa.py) Tunning script: ```bash BASE_MODEL=microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext OUTPUT_DIR=~/Documents/projects/tunned_models/ms_pubmed_bert_squadv2/ python run_qa.py \ --model_name_or_path $BASE_MODEL\ --dataset_name squad_v2 \ --do_train \ --do_eval \ --version_2_with_negative \ --per_device_train_batch_size 12 \ --learning_rate 3e-5 \ --num_train_epochs 2 \ --max_seq_length 384 \ --doc_stride 128 \ --output_dir $OUTPUT_DIR ```
fspanda/Medical-Bio-BERT2	f3fe25c06099883d3116d1395afa8df37fa0cb93	2021-05-19T16:57:41.000Z	[ "pytorch", "jax", "bert", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	fspanda	null	fspanda/Medical-Bio-BERT2	87	null	transformers	4,853	Entry not found
google/t5-large-ssm	0b0101c39a17cb38660e671692bdd892ba5d352d	2021-06-23T01:49:33.000Z	[ "pytorch", "tf", "jax", "t5", "text2text-generation", "en", "dataset:c4", "dataset:wikipedia", "arxiv:2002.08909", "arxiv:1910.10683", "transformers", "license:apache-2.0", "autotrain_compatible" ]	text2text-generation	false	google	null	google/t5-large-ssm	87	1	transformers	4,854	--- language: en datasets: - c4 - wikipedia license: apache-2.0 --- [Google's T5](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) for Closed Book Question Answering. The model was pre-trained using T5's denoising objective on [C4](https://huggingface.co/datasets/c4) and subsequently additionally pre-trained using [REALM](https://arxiv.org/pdf/2002.08909.pdf)'s salient span masking objective on [Wikipedia](https://huggingface.co/datasets/wikipedia). Note: This model should be fine-tuned on a question answering downstream task before it is useable for closed book question answering. Other Community Checkpoints: [here](https://huggingface.co/models?search=ssm) Paper: [How Much Knowledge Can You Pack Into the Parameters of a Language Model?](https://arxiv.org/abs/1910.10683.pdf) Authors: Adam Roberts, Colin Raffel, Noam Shazeer ## Abstract It has recently been observed that neural language models trained on unstructured text can implicitly store and retrieve knowledge using natural language queries. In this short paper, we measure the practical utility of this approach by fine-tuning pre-trained models to answer questions without access to any external context or knowledge. We show that this approach scales with model size and performs competitively with open-domain systems that explicitly retrieve answers from an external knowledge source when answering questions. To facilitate reproducibility and future work, we release our code and trained models at https://goo.gle/t5-cbqa. ![model image](https://raw.githubusercontent.com/patrickvonplaten/scientific_images/master/how_much_know_ledge_image.png)
illuin/camembert-base-fquad	8fea54f3caae6eef44ee1d1c52026ac48752a6c0	2020-12-11T21:45:27.000Z	[ "pytorch", "camembert", "question-answering", "fr", "dataset:fquad", "transformers", "license:gpl-3.0", "autotrain_compatible" ]	question-answering	false	illuin	null	illuin/camembert-base-fquad	87	3	transformers	4,855	--- language: fr tags: - question-answering - camembert license: gpl-3.0 datasets: - fquad --- # camembert-base-fquad ## Description A native French Question Answering model [CamemBERT-base](https://camembert-model.fr/) fine-tuned on [FQuAD](https://fquad.illuin.tech/). ## Evaluation results On the development set. ```shell {"f1": 88.1, "exact_match": 78.1} ``` On the test set. ```shell {"f1": 88.3, "exact_match": 78.0} ``` ## Usage ```python from transformers import pipeline nlp = pipeline('question-answering', model='illuin/camembert-base-fquad', tokenizer='illuin/camembert-base-fquad') nlp({ 'question': "Qui est Claude Monet?", 'context': "Claude Monet, né le 14 novembre 1840 à Paris et mort le 5 décembre 1926 à Giverny, est un peintre français et l’un des fondateurs de l'impressionnisme." }) ``` ## Citation If you use our work, please cite: ```bibtex @article{dHoffschmidt2020FQuADFQ, title={FQuAD: French Question Answering Dataset}, author={Martin d'Hoffschmidt and Maxime Vidal and Wacim Belblidia and Tom Brendl'e and Quentin Heinrich}, journal={ArXiv}, year={2020}, volume={abs/2002.06071} } ```
mrm8488/layoutlm-finetuned-funsd	3de04e17d2fb21729336dea31651b573e5e3c33a	2021-08-01T16:39:26.000Z	[ "pytorch", "layoutlm", "token-classification", "transformers", "autotrain_compatible" ]	token-classification	false	mrm8488	null	mrm8488/layoutlm-finetuned-funsd	87	null	transformers	4,856	# LayoutLM fine-tuned on FUNSD for Document/Forms token classification ## Usage (WIP) ```python import torch import numpy as np from PIL import Image, ImageDraw, ImageFont import pytesseract from transformers import LayoutLMForTokenClassification, LayoutLMTokenizer device = torch.device("cuda" if torch.cuda.is_available() else "cpu") tokenizer = LayoutLMTokenizer.from_pretrained("mrm8488/layoutlm-finetuned-funsd") model = LayoutLMForTokenClassification.from_pretrained("mrm8488/layoutlm-finetuned-funsd", num_labels=13) model.to(device) image = Image.open("/83443897.png") image = image.convert("RGB") # Display the image # Run Tesseract (OCR) on the image width, height = image.size w_scale = 1000/width h_scale = 1000/height ocr_df = pytesseract.image_to_data(image, output_type='data.frame') \\n ocr_df = ocr_df.dropna() \\n .assign(left_scaled = ocr_df.leftw_scale, width_scaled = ocr_df.widthw_scale, top_scaled = ocr_df.toph_scale, height_scaled = ocr_df.heighth_scale, right_scaled = lambda x: x.left_scaled + x.width_scaled, bottom_scaled = lambda x: x.top_scaled + x.height_scaled) float_cols = ocr_df.select_dtypes('float').columns ocr_df[float_cols] = ocr_df[float_cols].round(0).astype(int) ocr_df = ocr_df.replace(r'^\s$', np.nan, regex=True) ocr_df = ocr_df.dropna().reset_index(drop=True) ocr_df[:20] # create a list of words, actual bounding boxes, and normalized boxes words = list(ocr_df.text) coordinates = ocr_df[['left', 'top', 'width', 'height']] actual_boxes = [] for idx, row in coordinates.iterrows(): x, y, w, h = tuple(row) # the row comes in (left, top, width, height) format actual_box = [x, y, x+w, y+h] # we turn it into (left, top, left+widght, top+height) to get the actual box actual_boxes.append(actual_box) def normalize_box(box, width, height): return [ int(1000 (box[0] / width)), int(1000 * (box[1] / height)), int(1000 * (box[2] / width)), int(1000 * (box[3] / height)), ] boxes = [] for box in actual_boxes: boxes.append(normalize_box(box, width, height)) # Display boxes def convert_example_to_features(image, words, boxes, actual_boxes, tokenizer, args, cls_token_box=[0, 0, 0, 0], sep_token_box=[1000, 1000, 1000, 1000], pad_token_box=[0, 0, 0, 0]): width, height = image.size tokens = [] token_boxes = [] actual_bboxes = [] # we use an extra b because actual_boxes is already used token_actual_boxes = [] for word, box, actual_bbox in zip(words, boxes, actual_boxes): word_tokens = tokenizer.tokenize(word) tokens.extend(word_tokens) token_boxes.extend([box] * len(word_tokens)) actual_bboxes.extend([actual_bbox] * len(word_tokens)) token_actual_boxes.extend([actual_bbox] * len(word_tokens)) # Truncation: account for [CLS] and [SEP] with "- 2". special_tokens_count = 2 if len(tokens) > args.max_seq_length - special_tokens_count: tokens = tokens[: (args.max_seq_length - special_tokens_count)] token_boxes = token_boxes[: (args.max_seq_length - special_tokens_count)] actual_bboxes = actual_bboxes[: (args.max_seq_length - special_tokens_count)] token_actual_boxes = token_actual_boxes[: (args.max_seq_length - special_tokens_count)] # add [SEP] token, with corresponding token boxes and actual boxes tokens += [tokenizer.sep_token] token_boxes += [sep_token_box] actual_bboxes += [[0, 0, width, height]] token_actual_boxes += [[0, 0, width, height]] segment_ids = [0] * len(tokens) # next: [CLS] token tokens = [tokenizer.cls_token] + tokens token_boxes = [cls_token_box] + token_boxes actual_bboxes = [[0, 0, width, height]] + actual_bboxes token_actual_boxes = [[0, 0, width, height]] + token_actual_boxes segment_ids = [1] + segment_ids input_ids = tokenizer.convert_tokens_to_ids(tokens) # The mask has 1 for real tokens and 0 for padding tokens. Only real # tokens are attended to. input_mask = [1] * len(input_ids) # Zero-pad up to the sequence length. padding_length = args.max_seq_length - len(input_ids) input_ids += [tokenizer.pad_token_id] * padding_length input_mask += [0] * padding_length segment_ids += [tokenizer.pad_token_id] * padding_length token_boxes += [pad_token_box] * padding_length token_actual_boxes += [pad_token_box] * padding_length assert len(input_ids) == args.max_seq_length assert len(input_mask) == args.max_seq_length assert len(segment_ids) == args.max_seq_length assert len(token_boxes) == args.max_seq_length assert len(token_actual_boxes) == args.max_seq_length return input_ids, input_mask, segment_ids, token_boxes, token_actual_boxes input_ids, input_mask, segment_ids, token_boxes, token_actual_boxes = convert_example_to_features(image=image, words=words, boxes=boxes, actual_boxes=actual_boxes, tokenizer=tokenizer, args=args) input_ids = torch.tensor(input_ids, device=device).unsqueeze(0) attention_mask = torch.tensor(input_mask, device=device).unsqueeze(0) token_type_ids = torch.tensor(segment_ids, device=device).unsqueeze(0) bbox = torch.tensor(token_boxes, device=device).unsqueeze(0) outputs = model(input_ids=input_ids, bbox=bbox, attention_mask=attention_mask, token_type_ids=token_type_ids) token_predictions = outputs.logits.argmax(-1).squeeze().tolist() # the predictions are at the token level word_level_predictions = [] # let's turn them into word level predictions final_boxes = [] for id, token_pred, box in zip(input_ids.squeeze().tolist(), token_predictions, token_actual_boxes): if (tokenizer.decode([id]).startswith("##")) or (id in [tokenizer.cls_token_id, tokenizer.sep_token_id, tokenizer.pad_token_id]): # skip prediction + bounding box continue else: word_level_predictions.append(token_pred) final_boxes.append(box) #print(word_level_predictions) draw = ImageDraw.Draw(image) font = ImageFont.load_default() def iob_to_label(label): if label != 'O': return label[2:] else: return "other" label2color = {'question':'blue', 'answer':'green', 'header':'orange', 'other':'violet'} for prediction, box in zip(word_level_predictions, final_boxes): predicted_label = iob_to_label(label_map[prediction]).lower() draw.rectangle(box, outline=label2color[predicted_label]) draw.text((box[0] + 10, box[1] - 10), text=predicted_label, fill=label2color[predicted_label], font=font) # Display the result (image) ```
mrm8488/legalectra-small-spanish	4854a72623057dcf56e51f7efb3a0cb15398388e	2022-03-30T21:06:31.000Z	[ "pytorch", "electra", "pretraining", "es", "dataset:Spanish-legal-corpora", "transformers", "Spanish", "Electra", "Legal" ]	null	false	mrm8488	null	mrm8488/legalectra-small-spanish	87	2	transformers	4,857	--- language: es tags: - Spanish - Electra - Legal datasets: - Spanish-legal-corpora --- ## LEGALECTRA ⚖️ LEGALECTRA (small) is an Electra like model (discriminator in this case) trained on [A collection of corpora of Spanish legal domain](https://zenodo.org/record/5495529#.YZItp3vMLJw). As mentioned in the original [paper](https://openreview.net/pdf?id=r1xMH1BtvB): ELECTRA is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input tokens vs "fake" input tokens generated by another neural network, similar to the discriminator of a [GAN](https://arxiv.org/pdf/1406.2661.pdf). At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the [SQuAD 2.0](https://rajpurkar.github.io/SQuAD-explorer/) dataset. For a detailed description and experimental results, please refer the paper [ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators](https://openreview.net/pdf?id=r1xMH1BtvB). ## Training details The model was trained using the Electra base code for 3 days on 1 Tesla V100 16GB. ## Model details ⚙ \|Param\| # Value\| \|-----\|--------\| \|Layers\| 12 \| \|Hidden \| 256 \| \|Params\| 14M \| ## Evaluation metrics (for discriminator) 🧾 \|Metric \| # Score \| \|-------\|---------\| \|Accuracy\| 0.955\| \|Precision\| 0.790\| \|AUC \| 0.971\| ## Benchmarks 🔨 WIP 🚧 ## How to use the discriminator in `transformers` TBA ## Acknowledgments TBA ## Citation If you want to cite this model you can use this: ```bibtex @misc{mromero2022legalectra, title={Spanish Legal Electra (small)}, author={Romero, Manuel}, publisher={Hugging Face}, journal={Hugging Face Hub}, howpublished={\url{https://huggingface.co/mrm8488/legalectra-small-spanish}, year={2022} } ``` > Created by [Manuel Romero/@mrm8488](https://twitter.com/mrm8488) > Made with <span style="color: #e25555;">&hearts;</span> in Spain
nateraw/test_model_a	25e32086060f76efb5194dcb2db351b90ebf5981	2021-07-13T04:52:00.000Z	[ "pytorch", "vit", "image-classification", "dataset:image_folder", "transformers", "generated_from_trainer" ]	image-classification	false	nateraw	null	nateraw/test_model_a	87	null	transformers	4,858	--- tags: - generated_from_trainer datasets: - image_folder model_index: - name: test_model_a results: - task: name: Image Classification type: image-classification dataset: name: image_folder type: image_folder args: default --- <!-- 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. --> # test_model_a This model is a fine-tuned version of [lysandre/tiny-vit-random](https://huggingface.co/lysandre/tiny-vit-random) on the image_folder 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 - training_steps: 40 ### Framework versions - Transformers 4.8.2 - Pytorch 1.9.0+cu102 - Datasets 1.9.1.dev0 - Tokenizers 0.10.3
mustapha/flipped-image-ViT	3c76c3aa2325d0d5678c64e7b536ead6903e64b1	2022-03-31T12:30:19.000Z	[ "pytorch", "vit", "image-classification", "transformers" ]	image-classification	false	mustapha	null	mustapha/flipped-image-ViT	87	1	transformers	4,859	Hello world, This model have been created in the context of ` Fatima Fellowship Programme`. The model was trained on the Cifar10 dataset with a googd final accuracy of arround 98%. This model determines wether an image is flipped of not.
moshew/bert-mini-sst2-distilled	61728bcf9f705e4161ee2be3185bfb48f7a1c617	2022-04-13T11:33:29.000Z	[ "pytorch", "tensorboard", "bert", "text-classification", "transformers" ]	text-classification	false	moshew	null	moshew/bert-mini-sst2-distilled	87	null	transformers	4,860	Entry not found
tinkoff-ai/response-quality-classifier-base	4d3c0a32337d670038becdbd0478e85012795a3c	2022-06-01T06:34:22.000Z	[ "pytorch", "bert", "text-classification", "ru", "transformers", "conversational", "license:mit" ]	text-classification	false	tinkoff-ai	null	tinkoff-ai/response-quality-classifier-base	87	null	transformers	4,861	--- license: mit widget: - text: "[CLS]привет[SEP]привет![SEP]как дела?[RESPONSE_TOKEN]супер, вот только проснулся, у тебя как?" example_title: "Dialog example 1" - text: "[CLS]привет[SEP]привет![SEP]как дела?[RESPONSE_TOKEN]норм" example_title: "Dialog example 2" - text: "[CLS]привет[SEP]привет![SEP]как дела?[RESPONSE_TOKEN]норм, у тя как?" example_title: "Dialog example 3" language: - ru tags: - conversational --- This classification model is based on [DeepPavlov/rubert-base-cased-sentence](https://huggingface.co/DeepPavlov/rubert-base-cased-sentence). The model should be used to produce relevance and specificity of the last message in the context of a dialogue. The labels explanation: - `relevance`: is the last message in the dialogue relevant in the context of the full dialogue. - `specificity`: is the last message in the dialogue interesting and promotes the continuation of the dialogue. It is pretrained on a large corpus of dialog data in unsupervised manner: the model is trained to predict whether last response was in a real dialog, or it was pulled from some other dialog at random. Then it was finetuned on manually labelled examples (dataset will be posted soon). The model was trained with three messages in the context and one response. Each message was tokenized separately with ``` max_length = 32 ```. The performance of the model on validation split (dataset will be posted soon) (with the best thresholds for validation samples): \| \| threshold \| f0.5 \| ROC AUC \| \|:------------\|------------:\|-------:\|----------:\| \| relevance \| 0.49 \| 0.84 \| 0.79 \| \| specificity \| 0.53 \| 0.83 \| 0.83 \| How to use: ```python import torch from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained('tinkoff-ai/response-quality-classifier-base') model = AutoModelForSequenceClassification.from_pretrained('tinkoff-ai/response-quality-classifier-base') inputs = tokenizer('[CLS]привет[SEP]привет![SEP]как дела?[RESPONSE_TOKEN]норм, у тя как?', max_length=128, add_special_tokens=False, return_tensors='pt') with torch.inference_mode(): logits = model(**inputs).logits probas = torch.sigmoid(logits)[0].cpu().detach().numpy() relevance, specificity = probas ``` The [app](https://huggingface.co/spaces/tinkoff-ai/response-quality-classifiers) where you can easily interact with this model. The work was done during internship at Tinkoff by [egoriyaa](https://github.com/egoriyaa), mentored by [solemn-leader](https://huggingface.co/solemn-leader).
Theivaprakasham/layoutlmv3-finetuned-sroie	6c8e3d5dcdf9e36ca53ad490f93714c44bbce3a3	2022-06-07T18:08:04.000Z	[ "pytorch", "tensorboard", "layoutlmv3", "token-classification", "dataset:sroie", "transformers", "generated_from_trainer", "model-index", "autotrain_compatible" ]	token-classification	false	Theivaprakasham	null	Theivaprakasham/layoutlmv3-finetuned-sroie	87	null	transformers	4,862	--- tags: - generated_from_trainer datasets: - sroie metrics: - precision - recall - f1 - accuracy model-index: - name: layoutlmv3-finetuned-sroie results: - task: name: Token Classification type: token-classification dataset: name: sroie type: sroie args: sroie metrics: - name: Precision type: precision value: 0.9370529327610873 - name: Recall type: recall value: 0.9438040345821326 - name: F1 type: f1 value: 0.9404163675520459 - name: Accuracy type: accuracy value: 0.9945347083116948 --- <!-- 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. --> # layoutlmv3-finetuned-sroie This model is a fine-tuned version of [microsoft/layoutlmv3-base](https://huggingface.co/microsoft/layoutlmv3-base) on the sroie dataset. It achieves the following results on the evaluation set: - Loss: 0.0426 - Precision: 0.9371 - Recall: 0.9438 - F1: 0.9404 - Accuracy: 0.9945 ## 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: 2 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - training_steps: 5000 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Precision \| Recall \| F1 \| Accuracy \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:---------:\|:------:\|:------:\|:--------:\| \| No log \| 0.32 \| 100 \| 0.1127 \| 0.6466 \| 0.6102 \| 0.6279 \| 0.9729 \| \| No log \| 0.64 \| 200 \| 0.0663 \| 0.8215 \| 0.7428 \| 0.7802 \| 0.9821 \| \| No log \| 0.96 \| 300 \| 0.0563 \| 0.8051 \| 0.8718 \| 0.8371 \| 0.9855 \| \| No log \| 1.28 \| 400 \| 0.0470 \| 0.8766 \| 0.8595 \| 0.8680 \| 0.9895 \| \| 0.1328 \| 1.6 \| 500 \| 0.0419 \| 0.8613 \| 0.9128 \| 0.8863 \| 0.9906 \| \| 0.1328 \| 1.92 \| 600 \| 0.0338 \| 0.8888 \| 0.9099 \| 0.8993 \| 0.9926 \| \| 0.1328 \| 2.24 \| 700 \| 0.0320 \| 0.8690 \| 0.9467 \| 0.9062 \| 0.9929 \| \| 0.1328 \| 2.56 \| 800 \| 0.0348 \| 0.8960 \| 0.9438 \| 0.9193 \| 0.9931 \| \| 0.1328 \| 2.88 \| 900 \| 0.0300 \| 0.9169 \| 0.9460 \| 0.9312 \| 0.9942 \| \| 0.029 \| 3.19 \| 1000 \| 0.0281 \| 0.9080 \| 0.9452 \| 0.9262 \| 0.9942 \| \| 0.029 \| 3.51 \| 1100 \| 0.0259 \| 0.9174 \| 0.9438 \| 0.9304 \| 0.9945 \| \| 0.029 \| 3.83 \| 1200 \| 0.0309 \| 0.9207 \| 0.9532 \| 0.9366 \| 0.9944 \| \| 0.029 \| 4.15 \| 1300 \| 0.0366 \| 0.9195 \| 0.9388 \| 0.9291 \| 0.9940 \| \| 0.029 \| 4.47 \| 1400 \| 0.0302 \| 0.9343 \| 0.9424 \| 0.9383 \| 0.9949 \| \| 0.0174 \| 4.79 \| 1500 \| 0.0349 \| 0.9142 \| 0.9517 \| 0.9326 \| 0.9939 \| \| 0.0174 \| 5.11 \| 1600 \| 0.0327 \| 0.9322 \| 0.9510 \| 0.9415 \| 0.9950 \| \| 0.0174 \| 5.43 \| 1700 \| 0.0317 \| 0.9215 \| 0.9561 \| 0.9385 \| 0.9938 \| \| 0.0174 \| 5.75 \| 1800 \| 0.0385 \| 0.9282 \| 0.9316 \| 0.9299 \| 0.9940 \| \| 0.0174 \| 6.07 \| 1900 \| 0.0342 \| 0.9235 \| 0.9481 \| 0.9357 \| 0.9944 \| \| 0.0117 \| 6.39 \| 2000 \| 0.0344 \| 0.9287 \| 0.9474 \| 0.9379 \| 0.9944 \| \| 0.0117 \| 6.71 \| 2100 \| 0.0388 \| 0.9232 \| 0.9445 \| 0.9338 \| 0.9941 \| \| 0.0117 \| 7.03 \| 2200 \| 0.0325 \| 0.9269 \| 0.9496 \| 0.9381 \| 0.9949 \| \| 0.0117 \| 7.35 \| 2300 \| 0.0343 \| 0.9225 \| 0.9438 \| 0.9330 \| 0.9941 \| \| 0.0117 \| 7.67 \| 2400 \| 0.0372 \| 0.9216 \| 0.9481 \| 0.9347 \| 0.9944 \| \| 0.0081 \| 7.99 \| 2500 \| 0.0385 \| 0.9192 \| 0.9589 \| 0.9386 \| 0.9944 \| \| 0.0081 \| 8.31 \| 2600 \| 0.0376 \| 0.9293 \| 0.9467 \| 0.9379 \| 0.9944 \| \| 0.0081 \| 8.63 \| 2700 \| 0.0425 \| 0.9261 \| 0.9474 \| 0.9366 \| 0.9941 \| \| 0.0081 \| 8.95 \| 2800 \| 0.0407 \| 0.9266 \| 0.9452 \| 0.9358 \| 0.9941 \| \| 0.0081 \| 9.27 \| 2900 \| 0.0403 \| 0.9280 \| 0.9467 \| 0.9372 \| 0.9941 \| \| 0.0055 \| 9.58 \| 3000 \| 0.0364 \| 0.9287 \| 0.9474 \| 0.9379 \| 0.9948 \| \| 0.0055 \| 9.9 \| 3100 \| 0.0427 \| 0.9122 \| 0.9510 \| 0.9312 \| 0.9941 \| \| 0.0055 \| 10.22 \| 3200 \| 0.0394 \| 0.9223 \| 0.9488 \| 0.9354 \| 0.9943 \| \| 0.0055 \| 10.54 \| 3300 \| 0.0393 \| 0.9247 \| 0.9561 \| 0.9401 \| 0.9945 \| \| 0.0055 \| 10.86 \| 3400 \| 0.0413 \| 0.9334 \| 0.9496 \| 0.9414 \| 0.9945 \| \| 0.0049 \| 11.18 \| 3500 \| 0.0400 \| 0.9290 \| 0.9517 \| 0.9402 \| 0.9945 \| \| 0.0049 \| 11.5 \| 3600 \| 0.0412 \| 0.9317 \| 0.9539 \| 0.9427 \| 0.9945 \| \| 0.0049 \| 11.82 \| 3700 \| 0.0419 \| 0.9314 \| 0.9481 \| 0.9397 \| 0.9947 \| \| 0.0049 \| 12.14 \| 3800 \| 0.0452 \| 0.9243 \| 0.9503 \| 0.9371 \| 0.9941 \| \| 0.0049 \| 12.46 \| 3900 \| 0.0412 \| 0.9334 \| 0.9496 \| 0.9414 \| 0.9947 \| \| 0.0039 \| 12.78 \| 4000 \| 0.0438 \| 0.9294 \| 0.9481 \| 0.9387 \| 0.9941 \| \| 0.0039 \| 13.1 \| 4100 \| 0.0416 \| 0.9326 \| 0.9467 \| 0.9396 \| 0.9944 \| \| 0.0039 \| 13.42 \| 4200 \| 0.0418 \| 0.9327 \| 0.9488 \| 0.9407 \| 0.9948 \| \| 0.0039 \| 13.74 \| 4300 \| 0.0423 \| 0.9345 \| 0.9460 \| 0.9402 \| 0.9946 \| \| 0.0039 \| 14.06 \| 4400 \| 0.0419 \| 0.9286 \| 0.9467 \| 0.9376 \| 0.9947 \| \| 0.0022 \| 14.38 \| 4500 \| 0.0426 \| 0.9371 \| 0.9438 \| 0.9404 \| 0.9945 \| \| 0.0022 \| 14.7 \| 4600 \| 0.0424 \| 0.9371 \| 0.9445 \| 0.9408 \| 0.9947 \| \| 0.0022 \| 15.02 \| 4700 \| 0.0427 \| 0.9372 \| 0.9467 \| 0.9419 \| 0.9947 \| \| 0.0022 \| 15.34 \| 4800 \| 0.0431 \| 0.9339 \| 0.9460 \| 0.9399 \| 0.9945 \| \| 0.0022 \| 15.65 \| 4900 \| 0.0431 \| 0.9346 \| 0.9467 \| 0.9406 \| 0.9946 \| \| 0.0015 \| 15.97 \| 5000 \| 0.0434 \| 0.9324 \| 0.9445 \| 0.9384 \| 0.9945 \| ### Framework versions - Transformers 4.20.0.dev0 - Pytorch 1.11.0+cu113 - Datasets 2.2.2 - Tokenizers 0.12.1
crumb/gpt-j-6b-shakespeare	eee121823da348a498f0f544c6afdb34c7880923	2022-07-20T18:06:37.000Z	[ "pytorch", "gptj", "text-generation", "en", "dataset:The Pile", "dataset:tiny_shakespeare", "arxiv:2101.00027", "transformers", "causal-lm" ]	text-generation	false	crumb	null	crumb/gpt-j-6b-shakespeare	87	null	transformers	4,863	--- language: - en tags: - pytorch - causal-lm datasets: - The Pile - tiny_shakespeare inference: false --- # GPT-J 6b Shakespeare <p style="color:green"> <b> 1.) The "Hosted inference API" is turned off. Go to the <a href="https://huggingface.co/crumb/gpt-j-6b-shakespeare#how-to-use">How to Use</a> section <br> 2.) This is a "proof of concept" and not fully trained, simple training script also in "How to Use" section. </b> ## Model Description GPT-J 6B is a transformer model trained using Ben Wang's [Mesh Transformer JAX](https://github.com/kingoflolz/mesh-transformer-jax/). "GPT-J" refers to the class of model, while "6B" represents the number of trainable parameters. This checkpoint is a finetuned version of the original [GPT-J 6b](https://huggingface.co/EleutherAI/gpt-j-6B) on [tiny_shakespeare](https://huggingface.co/datasets/tiny_shakespeare) ## Training data GPT-J 6B was trained on [the Pile](https://pile.eleuther.ai), a large-scale curated dataset created by [EleutherAI](https://www.eleuther.ai). This checkpoint was afterwards finetuned on [tiny_shakespeare](https://huggingface.co/datasets/tiny_shakespeare) by [crumb](https://huggingface.co/crumb) (me) > 40,000 lines of Shakespeare from a variety of Shakespeare's plays. Featured in Andrej Karpathy's blog post 'The Unreasonable Effectiveness of Recurrent Neural Networks': http://karpathy.github.io/2015/05/21/rnn-effectiveness/. ## Training Procedure \| Parameter \| Value \| \|----------------------\|------------\| \| epochs \| 1 \| \| learning rate \| .002 \| \| weight decay \| .01 \| \| batch size \| 8 \| \| context length (tokens) \| 256 \| Trained on 1 Tesla T4 from [google colab](https://colab.research.google.com/) ```TrainOutput(global_step=147, training_loss=1.665000240818984, metrics={'train_runtime': 2828.7347, 'train_samples_per_second': 0.417, 'train_steps_per_second': 0.052, 'total_flos': 1555992281088.0, 'train_loss': 1.665000240818984, 'epoch': 1.0})``` A good starting point to finetune your own gpt-j-6b would be [hivemind's 8bit training code](https://huggingface.co/hivemind/gpt-j-6B-8bit), or with the notebook in [this repository](https://github.com/aicrumb/gpt-j-8bit) which you can download and open in [google colab](https://colab.research.google.com/) or any other ipynb service No LORA adapters were used for the sake of easy loading and inference with 🤗. Only Linear biases and LayerNorm scales were passed to the optimizer. ## Intended Use and Limitations (same as [gpt-j-6b](https://huggingface.co/EleutherAI/gpt-j-6B)) GPT-J learns an inner representation of the English language that can be used to extract features useful for downstream tasks. The model is best at what it was pretrained for however, which is generating text from a prompt. ### How to use ```python # libraries and a wrapper around hivemind's quantization code !pip install transformers==4.14.1 bitsandbytes-cuda111==0.26.0 git+https://github.com/aicrumb/transformers-8bit -q import transformers_8bit model, tokenizer, config = transformers_8bit.load_gptj("crumb/gpt-j-6b-shakespeare", device='cuda') prompt = tokenizer("Romeo:", return_tensors='pt') prompt = {key: value.to('cuda') for key, value in prompt.items()} out = model.generate(**prompt, min_length=64, max_length=64, do_sample=True, pad_token_id=tokenizer.eos_token_id) print(tokenizer.decode(out[0])) """ example output Romeo: [Aside] And but in night, how tedious Is the day's celebration! JULIET: [Aside] O me! how quick skips time! Bid Time himself look out And, after no long date, Call time up o'er-head, """ ``` ### Limitations and Biases (same as [gpt-j-6b](https://huggingface.co/EleutherAI/gpt-j-6B)) The core functionality of GPT-J is taking a string of text and predicting the next token. While language models are widely used for tasks other than this, there are a lot of unknowns with this work. When prompting GPT-J it is important to remember that the statistically most likely next token is often not the token that produces the most "accurate" text. Never depend upon GPT-J to produce factually accurate output. GPT-J was trained on the Pile, a dataset known to contain profanity, lewd, and otherwise abrasive language. Depending upon use case GPT-J may produce socially unacceptable text. See [Sections 5 and 6 of the Pile paper](https://arxiv.org/abs/2101.00027) for a more detailed analysis of the biases in the Pile. As with all language models, it is hard to predict in advance how GPT-J will respond to particular prompts and offensive content may occur without warning. We recommend having a human curate or filter the outputs before releasing them, both to censor undesirable content and to improve the quality of the results. ## To do: - clean up training code & create github repo for training related models - see if converting to fp16 or fp32 fixes the inference on the card ## Citations and Related Information ```bibtex @misc{gpt-j, author = {Wang, Ben and Komatsuzaki, Aran}, title = {{GPT-J-6B: A 6 Billion Parameter Autoregressive Language Model}}, howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}}, year = 2021, month = May } ``` ```bibtex @misc{mesh-transformer-jax, author = {Wang, Ben}, title = {{Mesh-Transformer-JAX: Model-Parallel Implementation of Transformer Language Model with JAX}}, howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}}, year = 2021, month = May } ``` ```bibtex @misc{ author={Karpathy, Andrej}, title={char-rnn}, year={2015}, howpublished={\url{https://github.com/karpathy/char-rnn}} } ```
tilomichel/mT5-base-GermanQuAD-e2e-qg	61f6573dde3826229f574d2aaf2ff4ea07d961cc	2022-07-02T10:46:35.000Z	[ "pytorch", "mt5", "text2text-generation", "de", "dataset:deepset/germanquad", "arxiv:2010.11934", "arxiv:2005.01107", "transformers", "question generation", "license:mit", "model-index", "autotrain_compatible" ]	text2text-generation	false	tilomichel	null	tilomichel/mT5-base-GermanQuAD-e2e-qg	87	null	transformers	4,864	--- license: mit widget: - text: "generate question: KMI ist eine Variante des allgemeinen Bachelors Informatik und damit zu ca. 80% identisch mit dem allgemeinen Bachelor Informatik, d.h. auch diese Variante ist ein Informatikstudium mit einem hohen Programmieranteil. Der Studienschwerpunkt adressiert insbesondere die heute geforderten Soft-Skills, die für ein Arbeiten im Team unerlässlich sind. Des Weiteren lernen Sie das Interaktionsdesign Ihrer Anwendungen kreativ zu optimieren und ihr Auge für eine gelungene Gestaltung zu schulen. In jedem Semester werden Akzente gesetzt: Im ersten und dritten Semester haben Sie beispielsweise ein Projekt anstelle eher technisch ausgerichteter Module. Die Hälfte Ihrer Wahlpflichtmodule absolvieren Sie am Fachbereich Media. </s>" example_title: "Question generation 1" - text: "generate question: SARS-CoV-2 zirkuliert weiterhin in der Bevölkerung und kann sich überall dort verbreiten, wo Menschen zusammenkommen. Auch wenn in den Sommermonaten die Fallzahlen saisonbedingt niedriger sind als in der kalten Jahreszeit, empfiehlt das RKI nach wie vor, die AHA+A+L-Regeln einzuhalten (Abstand halten, Hygieneregeln beachten, Alltag mit Maske, Coronawarnapp nutzen, Lüften), bei Atemwegssymptomen zu Hause zu bleiben und sich testen zu lassen, und auf einen vollständigen Impfschutz gegen COVID-19 zu achten. </s>" example_title: "Question generation 2" - text: "generate question: Ballaststoffe haben eine Reihe von Wirkungen auf den Körper, vor allem auf die Verdauung, z. B. Einfluss auf die Transitzeit der Nahrung in Magen und Darm, Masse und Konsistenz des Stuhls sowie Häufigkeit der Darmentleerung, Sättigungswirkung, veränderte Nährstoffabsorption und präbiotische Wirkung. Je nach Art der Ballaststoffe und nach Abschnitt im Verdauungstrakt kann es zu unterschiedlichen Effekten kommen. Bei der Fermentation von Ballaststoffen entstehen zudem verschiedene kurzkettige Fettsäuren, die dem Körper teilweise als Energiequelle zur Verfügung stehen. Schätzungsweise liefern die kurzkettigen Fettsäuren 8,4 kJ (2,0 kcal) pro g Ballaststoff. </s>" example_title: "Question generation 3" inference: parameters: max_length: 128 num_beams: 4 length_penalty: 1.5 no_repeat_ngram_size: 3 early_stopping: True language: - de tags: - question generation datasets: - deepset/germanquad metrics: - sacrebleu - bleu - rouge-l - meteor - bertscore model-index: - name: tilomichel/mT5-base-GermanQuAD-e2e-qg results: - task: type: question-generation name: Question generation dataset: type: xquad name: XQuAD (de) split: de metrics: - type: sacrebleu value: 1.72837804716791 name: BLEU Score args: lowercase: true verified: false - type: sacrebleu value: 49.210584834334 name: BLEU-1 args: lowercase: true verified: false - type: sacrebleu value: 16.960300681230915 name: BLEU-2 args: lowercase: true verified: false - type: sacrebleu value: 7.144635299975106 name: BLEU-3 args: lowercase: true verified: false - type: sacrebleu value: 3.230076780513635 name: BLEU-4 args: lowercase: true verified: false - type: rouge name: ROUGE-L (f-measure) value: 0.171130005590873 args: use_aggregator: true use_stemmer: false verified: false - type: meteor value: 0.0835049103331918 name: METEOR args: language: de verified: false - type: bertscore value: 0.331940584507538 name: BERTScore (F1) args: rescale_with_baseline: true verified: false --- # mT5-base finetuned on the GermanQuAD dataset for answer-agnostic question generation This model is a finetuned [mT5-base](https://arxiv.org/abs/2010.11934) model for the task of answer-agnostic (or end-to-end) question generation. The approach from [Lopez et al.](https://arxiv.org/abs/2005.01107) was used called All questions per line (AQPL). This means a paragraph is provided as input and multiple questions are generated from it. Other models already tested this approach with the T5 model for [English](https://huggingface.co/valhalla/t5-base-e2e-qg) and [German](https://huggingface.co/dehio/german-qg-t5-e2e-quad). For finetuning this model only used the [GermanQuAD dataset from deepset](https://www.deepset.ai/germanquad) was used. The dataset was modified and filtered with scripts that can be found in [another repository](https://github.com/TiloMichel/textgen-for-chatbot-training-german/tree/main/1_data_preparation_and_exploration). ## Training, test and evaluation data For training and test the original split from GermanQuAD was used. As evaluation dataset the German split of the [XQuAD](https://github.com/deepmind/xquad) dataset was used. ## Training hyperparameters The training parameters are provided in JSON and can be used with a training script provided in a [repository](https://github.com/TiloMichel/textgen-for-chatbot-training-german/tree/main/2_training) ```JS { "model_name_or_path": "google/mt5-base", "output_dir": "mt5-base-germanquad-e2e-qg", "overwrite_output_dir": true, "cache_dir": "model-cache", "dataset_dir": "e2e-qg-germanquad", "preprocessing_num_workers": 20, "max_source_length": 1024, "max_target_length": 128, "val_max_target_length": 128, "pad_to_max_length": true, "seed": 42, "do_train": true, "gradient_accumulation_steps": 64, "per_device_train_batch_size": 1, "per_device_eval_batch_size": 1, "learning_rate": 1e-4, "num_train_epochs": 10, "evaluation_strategy": "epoch", "logging_strategy": "epoch", "save_strategy": "epoch", "save_total_limit": 3, "dataloader_num_workers": 8, "ddp_find_unused_parameters": false } ``` ## Training results The evaluation is reported on XQuAD. The implementations and configurations can be found in [another repository](https://github.com/TiloMichel/textgen-for-chatbot-training-german/tree/main/3_evaluation).
natalierobbins/test_model	6229c433e4c8e29be7d43a783685b0345cf76bca	2022-07-19T23:29:56.000Z	[ "pytorch", "distilbert", "token-classification", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index", "autotrain_compatible" ]	token-classification	false	natalierobbins	null	natalierobbins/test_model	87	null	transformers	4,865	--- license: apache-2.0 tags: - generated_from_trainer metrics: - accuracy - f1 - precision - recall model-index: - name: test_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. --> # test_model 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: 0.0749 - Accuracy: 0.9720 - F1: 0.9698 - Precision: 0.9710 - Recall: 0.9720 ## 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: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Accuracy \| F1 \| Precision \| Recall \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:--------:\|:------:\|:---------:\|:------:\| \| 0.0812 \| 1.0 \| 1315 \| 0.0749 \| 0.9720 \| 0.9698 \| 0.9710 \| 0.9720 \| ### Framework versions - Transformers 4.18.0 - Pytorch 1.10.2 - Datasets 2.2.2 - Tokenizers 0.12.1
abelblue3/DialoGPT-medium-baymax	7e5216ccdbc9de4e8e2810f1bb38fadbcb8376de	2022-07-29T17:08:22.000Z	[ "pytorch", "gpt2", "text-generation", "transformers", "conversational" ]	conversational	false	abelblue3	null	abelblue3/DialoGPT-medium-baymax	87	null	transformers	4,866	--- tags: - conversational --- # DialoGPT BaymaxBot
AkshatSurolia/DeiT-FaceMask-Finetuned	7a7f81c61c64c9d0821d0a5d63d54fe9427ebde6	2022-02-18T13:10:05.000Z	[ "pytorch", "deit", "image-classification", "dataset:Face-Mask18K", "transformers", "license:apache-2.0" ]	image-classification	false	AkshatSurolia	null	AkshatSurolia/DeiT-FaceMask-Finetuned	86	null	transformers	4,867	--- license: apache-2.0 tags: - image-classification datasets: - Face-Mask18K --- # Distilled Data-efficient Image Transformer for Face Mask Detection Distilled data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on Self Currated Custom Face-Mask18K Dataset (18k images, 2 classes) at resolution 224x224. It was first introduced in the paper Training data-efficient image transformers & distillation through attention by Touvron et al. ## Model description This model is a distilled Vision Transformer (ViT). It uses a distillation token, besides the class token, to effectively learn from a teacher (CNN) during both pre-training and fine-tuning. The distillation token is learned through backpropagation, by interacting with the class ([CLS]) and patch tokens through the self-attention layers. Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. ## Training Metrics epoch = 2.0 total_flos = 2078245655GF train_loss = 0.0438 train_runtime = 1:37:16.87 train_samples_per_second = 9.887 train_steps_per_second = 0.309 --- ## Evaluation Metrics epoch = 2.0 eval_accuracy = 0.9922 eval_loss = 0.0271 eval_runtime = 0:03:17.36 eval_samples_per_second = 18.22 eval_steps_per_second = 2.28
BigSalmon/SimplifyText	f1dcd3f3a42cf9f8515c8177e341c506e24dd9d2	2021-10-14T00:41:11.000Z	[ "pytorch", "gpt2", "text-generation", "transformers" ]	text-generation	false	BigSalmon	null	BigSalmon/SimplifyText	86	null	transformers	4,868	- All credit goes to https://huggingface.co/philippelaban/keep_it_simple. - This is a copy of their repository for future training purposes. - It is supposed to simplify text. - Their model card gives instructions on how to use it.
Geotrend/bert-base-en-th-cased	336fd71f6da084fd96bc5cec3b745712333ba686	2021-05-18T19:47:11.000Z	[ "pytorch", "tf", "jax", "bert", "fill-mask", "multilingual", "dataset:wikipedia", "transformers", "license:apache-2.0", "autotrain_compatible" ]	fill-mask	false	Geotrend	null	Geotrend/bert-base-en-th-cased	86	null	transformers	4,869	--- language: multilingual datasets: wikipedia license: apache-2.0 widget: - text: "Google generated 46 billion [MASK] in revenue." - text: "Paris is the capital of [MASK]." - text: "Algiers is the largest city in [MASK]." --- # bert-base-en-th-cased We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages. Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy. For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf). ## How to use ```python from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-th-cased") model = AutoModel.from_pretrained("Geotrend/bert-base-en-th-cased") ``` To generate other smaller versions of multilingual transformers please visit [our Github repo](https://github.com/Geotrend-research/smaller-transformers). ### How to cite ```bibtex @inproceedings{smallermbert, title={Load What You Need: Smaller Versions of Mutlilingual BERT}, author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire}, booktitle={SustaiNLP / EMNLP}, year={2020} } ``` ## Contact Please contact [email protected] for any question, feedback or request.
Helsinki-NLP/opus-mt-de-vi	ec56a3fd05ff4fba8acaa11a0d434e889b088cdd	2021-01-18T08:02:50.000Z	[ "pytorch", "marian", "text2text-generation", "de", "vi", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-de-vi	86	null	transformers	4,870	--- language: - de - vi tags: - translation license: apache-2.0 --- ### deu-vie * source group: German * target group: Vietnamese * OPUS readme: [deu-vie](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/deu-vie/README.md) * model: transformer-align * source language(s): deu * target language(s): vie * model: transformer-align * pre-processing: normalization + SentencePiece (spm32k,spm32k) * download original weights: [opus-2020-06-17.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/deu-vie/opus-2020-06-17.zip) * test set translations: [opus-2020-06-17.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/deu-vie/opus-2020-06-17.test.txt) * test set scores: [opus-2020-06-17.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/deu-vie/opus-2020-06-17.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba-test.deu.vie \| 25.0 \| 0.443 \| ### System Info: - hf_name: deu-vie - source_languages: deu - target_languages: vie - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/deu-vie/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['de', 'vi'] - src_constituents: {'deu'} - tgt_constituents: {'vie', 'vie_Hani'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/deu-vie/opus-2020-06-17.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/deu-vie/opus-2020-06-17.test.txt - src_alpha3: deu - tgt_alpha3: vie - short_pair: de-vi - chrF2_score: 0.44299999999999995 - bleu: 25.0 - brevity_penalty: 1.0 - ref_len: 3768.0 - src_name: German - tgt_name: Vietnamese - train_date: 2020-06-17 - src_alpha2: de - tgt_alpha2: vi - prefer_old: False - long_pair: deu-vie - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
Helsinki-NLP/opus-mt-gmw-gmw	14861fa7cfac10a6f99c09e30a18c23cc02bf1b1	2021-01-18T08:53:04.000Z	[ "pytorch", "marian", "text2text-generation", "nl", "en", "lb", "af", "de", "fy", "yi", "gmw", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-gmw-gmw	86	null	transformers	4,871	--- language: - nl - en - lb - af - de - fy - yi - gmw tags: - translation license: apache-2.0 --- ### gmw-gmw * source group: West Germanic languages * target group: West Germanic languages * OPUS readme: [gmw-gmw](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/gmw-gmw/README.md) * model: transformer * source language(s): afr ang_Latn deu eng enm_Latn frr fry gos gsw ksh ltz nds nld pdc sco stq swg yid * target language(s): afr ang_Latn deu eng enm_Latn frr fry gos gsw ksh ltz nds nld pdc sco stq swg yid * model: transformer * pre-processing: normalization + SentencePiece (spm32k,spm32k) * a sentence initial language token is required in the form of `>>id<<` (id = valid target language ID) * download original weights: [opus-2020-07-27.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/gmw-gmw/opus-2020-07-27.zip) * test set translations: [opus-2020-07-27.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/gmw-gmw/opus-2020-07-27.test.txt) * test set scores: [opus-2020-07-27.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/gmw-gmw/opus-2020-07-27.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| newssyscomb2009-deueng.deu.eng \| 25.3 \| 0.527 \| \| newssyscomb2009-engdeu.eng.deu \| 19.0 \| 0.502 \| \| news-test2008-deueng.deu.eng \| 23.7 \| 0.515 \| \| news-test2008-engdeu.eng.deu \| 19.2 \| 0.491 \| \| newstest2009-deueng.deu.eng \| 23.1 \| 0.514 \| \| newstest2009-engdeu.eng.deu \| 18.6 \| 0.495 \| \| newstest2010-deueng.deu.eng \| 25.8 \| 0.545 \| \| newstest2010-engdeu.eng.deu \| 20.3 \| 0.505 \| \| newstest2011-deueng.deu.eng \| 23.7 \| 0.523 \| \| newstest2011-engdeu.eng.deu \| 18.9 \| 0.490 \| \| newstest2012-deueng.deu.eng \| 24.4 \| 0.529 \| \| newstest2012-engdeu.eng.deu \| 19.2 \| 0.489 \| \| newstest2013-deueng.deu.eng \| 27.2 \| 0.545 \| \| newstest2013-engdeu.eng.deu \| 22.4 \| 0.514 \| \| newstest2014-deen-deueng.deu.eng \| 27.0 \| 0.546 \| \| newstest2015-ende-deueng.deu.eng \| 28.4 \| 0.552 \| \| newstest2015-ende-engdeu.eng.deu \| 25.3 \| 0.541 \| \| newstest2016-ende-deueng.deu.eng \| 33.2 \| 0.595 \| \| newstest2016-ende-engdeu.eng.deu \| 29.8 \| 0.578 \| \| newstest2017-ende-deueng.deu.eng \| 29.0 \| 0.557 \| \| newstest2017-ende-engdeu.eng.deu \| 23.9 \| 0.534 \| \| newstest2018-ende-deueng.deu.eng \| 35.9 \| 0.607 \| \| newstest2018-ende-engdeu.eng.deu \| 34.8 \| 0.609 \| \| newstest2019-deen-deueng.deu.eng \| 32.1 \| 0.579 \| \| newstest2019-ende-engdeu.eng.deu \| 31.0 \| 0.579 \| \| Tatoeba-test.afr-ang.afr.ang \| 0.0 \| 0.065 \| \| Tatoeba-test.afr-deu.afr.deu \| 46.8 \| 0.668 \| \| Tatoeba-test.afr-eng.afr.eng \| 58.5 \| 0.728 \| \| Tatoeba-test.afr-enm.afr.enm \| 13.4 \| 0.357 \| \| Tatoeba-test.afr-fry.afr.fry \| 5.3 \| 0.026 \| \| Tatoeba-test.afr-gos.afr.gos \| 3.5 \| 0.228 \| \| Tatoeba-test.afr-ltz.afr.ltz \| 1.6 \| 0.131 \| \| Tatoeba-test.afr-nld.afr.nld \| 55.4 \| 0.715 \| \| Tatoeba-test.afr-yid.afr.yid \| 3.4 \| 0.008 \| \| Tatoeba-test.ang-afr.ang.afr \| 3.1 \| 0.096 \| \| Tatoeba-test.ang-deu.ang.deu \| 2.6 \| 0.188 \| \| Tatoeba-test.ang-eng.ang.eng \| 5.4 \| 0.211 \| \| Tatoeba-test.ang-enm.ang.enm \| 1.7 \| 0.197 \| \| Tatoeba-test.ang-gos.ang.gos \| 6.6 \| 0.186 \| \| Tatoeba-test.ang-ltz.ang.ltz \| 5.3 \| 0.072 \| \| Tatoeba-test.ang-yid.ang.yid \| 0.9 \| 0.131 \| \| Tatoeba-test.deu-afr.deu.afr \| 52.7 \| 0.699 \| \| Tatoeba-test.deu-ang.deu.ang \| 0.8 \| 0.133 \| \| Tatoeba-test.deu-eng.deu.eng \| 43.5 \| 0.621 \| \| Tatoeba-test.deu-enm.deu.enm \| 6.9 \| 0.245 \| \| Tatoeba-test.deu-frr.deu.frr \| 0.8 \| 0.200 \| \| Tatoeba-test.deu-fry.deu.fry \| 15.1 \| 0.367 \| \| Tatoeba-test.deu-gos.deu.gos \| 2.2 \| 0.279 \| \| Tatoeba-test.deu-gsw.deu.gsw \| 1.0 \| 0.176 \| \| Tatoeba-test.deu-ksh.deu.ksh \| 0.6 \| 0.208 \| \| Tatoeba-test.deu-ltz.deu.ltz \| 12.1 \| 0.274 \| \| Tatoeba-test.deu-nds.deu.nds \| 18.8 \| 0.446 \| \| Tatoeba-test.deu-nld.deu.nld \| 48.6 \| 0.669 \| \| Tatoeba-test.deu-pdc.deu.pdc \| 4.6 \| 0.198 \| \| Tatoeba-test.deu-sco.deu.sco \| 12.0 \| 0.340 \| \| Tatoeba-test.deu-stq.deu.stq \| 3.2 \| 0.240 \| \| Tatoeba-test.deu-swg.deu.swg \| 0.5 \| 0.179 \| \| Tatoeba-test.deu-yid.deu.yid \| 1.7 \| 0.160 \| \| Tatoeba-test.eng-afr.eng.afr \| 55.8 \| 0.730 \| \| Tatoeba-test.eng-ang.eng.ang \| 5.7 \| 0.157 \| \| Tatoeba-test.eng-deu.eng.deu \| 36.7 \| 0.584 \| \| Tatoeba-test.eng-enm.eng.enm \| 2.0 \| 0.272 \| \| Tatoeba-test.eng-frr.eng.frr \| 6.1 \| 0.246 \| \| Tatoeba-test.eng-fry.eng.fry \| 15.3 \| 0.378 \| \| Tatoeba-test.eng-gos.eng.gos \| 1.2 \| 0.242 \| \| Tatoeba-test.eng-gsw.eng.gsw \| 0.9 \| 0.164 \| \| Tatoeba-test.eng-ksh.eng.ksh \| 0.9 \| 0.170 \| \| Tatoeba-test.eng-ltz.eng.ltz \| 13.7 \| 0.263 \| \| Tatoeba-test.eng-nds.eng.nds \| 17.1 \| 0.410 \| \| Tatoeba-test.eng-nld.eng.nld \| 49.6 \| 0.673 \| \| Tatoeba-test.eng-pdc.eng.pdc \| 5.1 \| 0.218 \| \| Tatoeba-test.eng-sco.eng.sco \| 34.8 \| 0.587 \| \| Tatoeba-test.eng-stq.eng.stq \| 2.1 \| 0.322 \| \| Tatoeba-test.eng-swg.eng.swg \| 1.7 \| 0.192 \| \| Tatoeba-test.eng-yid.eng.yid \| 1.7 \| 0.173 \| \| Tatoeba-test.enm-afr.enm.afr \| 13.4 \| 0.397 \| \| Tatoeba-test.enm-ang.enm.ang \| 0.7 \| 0.063 \| \| Tatoeba-test.enm-deu.enm.deu \| 41.5 \| 0.514 \| \| Tatoeba-test.enm-eng.enm.eng \| 21.3 \| 0.483 \| \| Tatoeba-test.enm-fry.enm.fry \| 0.0 \| 0.058 \| \| Tatoeba-test.enm-gos.enm.gos \| 10.7 \| 0.354 \| \| Tatoeba-test.enm-ksh.enm.ksh \| 7.0 \| 0.161 \| \| Tatoeba-test.enm-nds.enm.nds \| 18.6 \| 0.316 \| \| Tatoeba-test.enm-nld.enm.nld \| 38.3 \| 0.524 \| \| Tatoeba-test.enm-yid.enm.yid \| 0.7 \| 0.128 \| \| Tatoeba-test.frr-deu.frr.deu \| 4.1 \| 0.219 \| \| Tatoeba-test.frr-eng.frr.eng \| 14.1 \| 0.186 \| \| Tatoeba-test.frr-fry.frr.fry \| 3.1 \| 0.129 \| \| Tatoeba-test.frr-gos.frr.gos \| 3.6 \| 0.226 \| \| Tatoeba-test.frr-nds.frr.nds \| 12.4 \| 0.145 \| \| Tatoeba-test.frr-nld.frr.nld \| 9.8 \| 0.209 \| \| Tatoeba-test.frr-stq.frr.stq \| 2.8 \| 0.142 \| \| Tatoeba-test.fry-afr.fry.afr \| 0.0 \| 1.000 \| \| Tatoeba-test.fry-deu.fry.deu \| 30.1 \| 0.535 \| \| Tatoeba-test.fry-eng.fry.eng \| 28.0 \| 0.486 \| \| Tatoeba-test.fry-enm.fry.enm \| 16.0 \| 0.262 \| \| Tatoeba-test.fry-frr.fry.frr \| 5.5 \| 0.160 \| \| Tatoeba-test.fry-gos.fry.gos \| 1.6 \| 0.307 \| \| Tatoeba-test.fry-ltz.fry.ltz \| 30.4 \| 0.438 \| \| Tatoeba-test.fry-nds.fry.nds \| 8.1 \| 0.083 \| \| Tatoeba-test.fry-nld.fry.nld \| 41.4 \| 0.616 \| \| Tatoeba-test.fry-stq.fry.stq \| 1.6 \| 0.217 \| \| Tatoeba-test.fry-yid.fry.yid \| 1.6 \| 0.159 \| \| Tatoeba-test.gos-afr.gos.afr \| 6.3 \| 0.318 \| \| Tatoeba-test.gos-ang.gos.ang \| 6.2 \| 0.058 \| \| Tatoeba-test.gos-deu.gos.deu \| 11.7 \| 0.363 \| \| Tatoeba-test.gos-eng.gos.eng \| 14.9 \| 0.322 \| \| Tatoeba-test.gos-enm.gos.enm \| 9.1 \| 0.398 \| \| Tatoeba-test.gos-frr.gos.frr \| 3.3 \| 0.117 \| \| Tatoeba-test.gos-fry.gos.fry \| 13.1 \| 0.387 \| \| Tatoeba-test.gos-ltz.gos.ltz \| 3.1 \| 0.154 \| \| Tatoeba-test.gos-nds.gos.nds \| 2.4 \| 0.206 \| \| Tatoeba-test.gos-nld.gos.nld \| 13.9 \| 0.395 \| \| Tatoeba-test.gos-stq.gos.stq \| 2.1 \| 0.209 \| \| Tatoeba-test.gos-yid.gos.yid \| 1.7 \| 0.147 \| \| Tatoeba-test.gsw-deu.gsw.deu \| 10.5 \| 0.350 \| \| Tatoeba-test.gsw-eng.gsw.eng \| 10.7 \| 0.299 \| \| Tatoeba-test.ksh-deu.ksh.deu \| 12.0 \| 0.373 \| \| Tatoeba-test.ksh-eng.ksh.eng \| 3.2 \| 0.225 \| \| Tatoeba-test.ksh-enm.ksh.enm \| 13.4 \| 0.308 \| \| Tatoeba-test.ltz-afr.ltz.afr \| 37.4 \| 0.525 \| \| Tatoeba-test.ltz-ang.ltz.ang \| 2.8 \| 0.036 \| \| Tatoeba-test.ltz-deu.ltz.deu \| 40.3 \| 0.596 \| \| Tatoeba-test.ltz-eng.ltz.eng \| 31.7 \| 0.490 \| \| Tatoeba-test.ltz-fry.ltz.fry \| 36.3 \| 0.658 \| \| Tatoeba-test.ltz-gos.ltz.gos \| 2.9 \| 0.209 \| \| Tatoeba-test.ltz-nld.ltz.nld \| 38.8 \| 0.530 \| \| Tatoeba-test.ltz-stq.ltz.stq \| 5.8 \| 0.165 \| \| Tatoeba-test.ltz-yid.ltz.yid \| 1.0 \| 0.159 \| \| Tatoeba-test.multi.multi \| 36.4 \| 0.568 \| \| Tatoeba-test.nds-deu.nds.deu \| 35.0 \| 0.573 \| \| Tatoeba-test.nds-eng.nds.eng \| 29.6 \| 0.495 \| \| Tatoeba-test.nds-enm.nds.enm \| 3.7 \| 0.194 \| \| Tatoeba-test.nds-frr.nds.frr \| 6.6 \| 0.133 \| \| Tatoeba-test.nds-fry.nds.fry \| 4.2 \| 0.087 \| \| Tatoeba-test.nds-gos.nds.gos \| 2.0 \| 0.243 \| \| Tatoeba-test.nds-nld.nds.nld \| 41.4 \| 0.618 \| \| Tatoeba-test.nds-swg.nds.swg \| 0.6 \| 0.178 \| \| Tatoeba-test.nds-yid.nds.yid \| 8.3 \| 0.238 \| \| Tatoeba-test.nld-afr.nld.afr \| 59.4 \| 0.759 \| \| Tatoeba-test.nld-deu.nld.deu \| 49.9 \| 0.685 \| \| Tatoeba-test.nld-eng.nld.eng \| 54.1 \| 0.699 \| \| Tatoeba-test.nld-enm.nld.enm \| 5.0 \| 0.250 \| \| Tatoeba-test.nld-frr.nld.frr \| 2.4 \| 0.224 \| \| Tatoeba-test.nld-fry.nld.fry \| 19.4 \| 0.446 \| \| Tatoeba-test.nld-gos.nld.gos \| 2.5 \| 0.273 \| \| Tatoeba-test.nld-ltz.nld.ltz \| 13.8 \| 0.292 \| \| Tatoeba-test.nld-nds.nld.nds \| 21.3 \| 0.457 \| \| Tatoeba-test.nld-sco.nld.sco \| 14.7 \| 0.423 \| \| Tatoeba-test.nld-stq.nld.stq \| 1.9 \| 0.257 \| \| Tatoeba-test.nld-swg.nld.swg \| 4.2 \| 0.162 \| \| Tatoeba-test.nld-yid.nld.yid \| 2.6 \| 0.186 \| \| Tatoeba-test.pdc-deu.pdc.deu \| 39.7 \| 0.529 \| \| Tatoeba-test.pdc-eng.pdc.eng \| 25.0 \| 0.427 \| \| Tatoeba-test.sco-deu.sco.deu \| 28.4 \| 0.428 \| \| Tatoeba-test.sco-eng.sco.eng \| 41.8 \| 0.595 \| \| Tatoeba-test.sco-nld.sco.nld \| 36.4 \| 0.565 \| \| Tatoeba-test.stq-deu.stq.deu \| 7.7 \| 0.328 \| \| Tatoeba-test.stq-eng.stq.eng \| 21.1 \| 0.428 \| \| Tatoeba-test.stq-frr.stq.frr \| 2.0 \| 0.118 \| \| Tatoeba-test.stq-fry.stq.fry \| 6.3 \| 0.255 \| \| Tatoeba-test.stq-gos.stq.gos \| 1.4 \| 0.244 \| \| Tatoeba-test.stq-ltz.stq.ltz \| 4.4 \| 0.204 \| \| Tatoeba-test.stq-nld.stq.nld \| 10.7 \| 0.371 \| \| Tatoeba-test.stq-yid.stq.yid \| 1.4 \| 0.105 \| \| Tatoeba-test.swg-deu.swg.deu \| 9.5 \| 0.343 \| \| Tatoeba-test.swg-eng.swg.eng \| 15.1 \| 0.306 \| \| Tatoeba-test.swg-nds.swg.nds \| 0.7 \| 0.196 \| \| Tatoeba-test.swg-nld.swg.nld \| 11.6 \| 0.308 \| \| Tatoeba-test.swg-yid.swg.yid \| 0.9 \| 0.186 \| \| Tatoeba-test.yid-afr.yid.afr \| 100.0 \| 1.000 \| \| Tatoeba-test.yid-ang.yid.ang \| 0.6 \| 0.079 \| \| Tatoeba-test.yid-deu.yid.deu \| 16.7 \| 0.372 \| \| Tatoeba-test.yid-eng.yid.eng \| 15.8 \| 0.344 \| \| Tatoeba-test.yid-enm.yid.enm \| 1.3 \| 0.166 \| \| Tatoeba-test.yid-fry.yid.fry \| 5.6 \| 0.157 \| \| Tatoeba-test.yid-gos.yid.gos \| 2.2 \| 0.160 \| \| Tatoeba-test.yid-ltz.yid.ltz \| 2.1 \| 0.238 \| \| Tatoeba-test.yid-nds.yid.nds \| 14.4 \| 0.365 \| \| Tatoeba-test.yid-nld.yid.nld \| 20.9 \| 0.397 \| \| Tatoeba-test.yid-stq.yid.stq \| 3.7 \| 0.165 \| \| Tatoeba-test.yid-swg.yid.swg \| 1.8 \| 0.156 \| ### System Info: - hf_name: gmw-gmw - source_languages: gmw - target_languages: gmw - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/gmw-gmw/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['nl', 'en', 'lb', 'af', 'de', 'fy', 'yi', 'gmw'] - src_constituents: {'ksh', 'nld', 'eng', 'enm_Latn', 'ltz', 'stq', 'afr', 'pdc', 'deu', 'gos', 'ang_Latn', 'fry', 'gsw', 'frr', 'nds', 'yid', 'swg', 'sco'} - tgt_constituents: {'ksh', 'nld', 'eng', 'enm_Latn', 'ltz', 'stq', 'afr', 'pdc', 'deu', 'gos', 'ang_Latn', 'fry', 'gsw', 'frr', 'nds', 'yid', 'swg', 'sco'} - src_multilingual: True - tgt_multilingual: True - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/gmw-gmw/opus-2020-07-27.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/gmw-gmw/opus-2020-07-27.test.txt - src_alpha3: gmw - tgt_alpha3: gmw - short_pair: gmw-gmw - chrF2_score: 0.568 - bleu: 36.4 - brevity_penalty: 1.0 - ref_len: 72534.0 - src_name: West Germanic languages - tgt_name: West Germanic languages - train_date: 2020-07-27 - src_alpha2: gmw - tgt_alpha2: gmw - prefer_old: False - long_pair: gmw-gmw - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
Helsinki-NLP/opus-mt-ja-tr	ac9ece2f2c4dd604a6d69c927bef045f09295e5d	2020-08-21T14:42:47.000Z	[ "pytorch", "marian", "text2text-generation", "ja", "tr", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-ja-tr	86	null	transformers	4,872	--- language: - ja - tr tags: - translation license: apache-2.0 --- ### jpn-tur * source group: Japanese * target group: Turkish * OPUS readme: [jpn-tur](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/jpn-tur/README.md) * model: transformer-align * source language(s): jpn jpn_Bopo jpn_Hang jpn_Hani jpn_Hira jpn_Kana jpn_Yiii * target language(s): tur * model: transformer-align * pre-processing: normalization + SentencePiece (spm32k,spm32k) * download original weights: [opus-2020-06-17.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-tur/opus-2020-06-17.zip) * test set translations: [opus-2020-06-17.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-tur/opus-2020-06-17.test.txt) * test set scores: [opus-2020-06-17.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-tur/opus-2020-06-17.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba-test.jpn.tur \| 16.7 \| 0.434 \| ### System Info: - hf_name: jpn-tur - source_languages: jpn - target_languages: tur - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/jpn-tur/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['ja', 'tr'] - src_constituents: {'jpn_Hang', 'jpn', 'jpn_Yiii', 'jpn_Kana', 'jpn_Hani', 'jpn_Bopo', 'jpn_Latn', 'jpn_Hira'} - tgt_constituents: {'tur'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-tur/opus-2020-06-17.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-tur/opus-2020-06-17.test.txt - src_alpha3: jpn - tgt_alpha3: tur - short_pair: ja-tr - chrF2_score: 0.434 - bleu: 16.7 - brevity_penalty: 0.932 - ref_len: 4755.0 - src_name: Japanese - tgt_name: Turkish - train_date: 2020-06-17 - src_alpha2: ja - tgt_alpha2: tr - prefer_old: False - long_pair: jpn-tur - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
NYTK/summarization-hi-bart-base-1024-hungarian	89a30ed7fa730cd8068cf4931de9fd4b41b21334	2022-02-14T13:27:40.000Z	[ "pytorch", "bart", "text2text-generation", "hu", "transformers", "summarization", "license:gpl", "autotrain_compatible" ]	summarization	false	NYTK	null	NYTK/summarization-hi-bart-base-1024-hungarian	86	null	transformers	4,873	--- language: - hu tags: - summarization license: gpl metrics: - rouge widget: - text: "A Tisza-parti város állatkertjében régóta tartanak szurikátákat ( Suricata suricatta ) , de tavaly tavaszig nem sikerült szaporítani őket , annak ellenére , hogy tágas ház és kifutó épült számukra - közölte Veprik Róbert igazgató . 2010-ben alakult ki az új - három Amszterdamból származó nőstényből és egy budapesti fiatal hímből álló - csapat , amely szaporodni kezdett . 2011-ben három , idén pedig egy utóddal örvendeztették meg a gondozókat és az állatbarátokat . A szurikáták utódai - tizenegy hetes vemhesség után - október és március között vakon és szőrtelenül jönnek a világra . A kicsinyek háromhetesen bújnak elő az üregből , és nevelésükben mindkét szülő részt vesz . A szurikátacsapatokban a család tagjai nagyon szoros kapcsolatban állnak egymással , viszont nagyon harciasan fellépnek az idegenekkel szemben , akár meg is ölhetik azt az állatot , amelyet betolakodónak tekintenek . Bár a Dél-Afrikában , a Kalahári sivatagban őshonos cibetmacskaféle ragadozókat a szegedi állatkertben természetes élőhelyükhöz képest kevesebb veszély fenyegeti , a vadasparki erdőben ragadozó madarak is élnek , amelyek akár zsákmányként is tekinthetnének a szurikátákra . A szegedi csapatnál azonban szigorú őrség van , mindig lesi valaki két lábra állva a veszélyforrásokat . Az őrszemek figyelmét még a sárkányrepülők is felkeltik , és felbukkanásakor valamennyi egyed biztos helyre menekül . A szurikáták a Kalahári sivatag bozótos , sziklás területein csapatokban élnek . A 700 gramm körüli testtömegű ragadozók rovarokkal , lárvákkal , skorpiókkal táplálkoznak , de néha elfogyasztják a kisebb gerinceseket , tojásokat és növényi gumókat is . A nappal aktív állatok földalatti üregrendszert ásnak , amelynek több bejárata is van . Ha a szurikáták idegen csapattal vagy ragadozóval kerülnek szembe , azonnal elkezdenek ásni , nagy porfelhőt kavarva . Az is gyakorta előfordul , hogy szorosan egymáshoz bújnak , felborzolják szőrüket , megnyújtják testüket , hogy minél nagyobbnak látszódjanak . Az előadásuk csúcspontján pedig az egész csapat a levegőbe ugrik , közben pedig morog . A hangadás egyébként is fontos a szurikáták kapcsolatában , az egyedek legalább tízféle jelzést használnak a kolónián belül ." --- # Hungarian Abstractive Summarization BART model For further models, scripts and details, see [our repository](https://github.com/nytud/neural-models) or [our demo site](https://juniper.nytud.hu/demo/nlp). - BART base model (see Results Table - bold): - Pretrained on Webcorpus 2.0 - Finetuned HI corpus (hvg.hu + index.hu) - Segments: 559.162 ## Limitations - tokenized input text (tokenizer: [HuSpaCy](https://huggingface.co/huspacy)) - max_source_length = 1024 - max_target_length = 256 ## Results \| Model \| HI \| NOL \| \| ------------- \| ------------- \| ------------- \| \| BART-base-512 \| 30.18/13.86/22.92 \| 46.48/32.40/39.45 \| \| BART-base-1024\| 31.86/14.59/23.79 \| 47.01/32.91/39.97 \| ## Citation If you use this model, please cite the following paper: ``` @inproceedings {yang-bart, title = {{BARTerezzünk! - Messze, messze, messze a világtól, - BART kísérleti modellek magyar nyelvre}}, booktitle = {XVIII. Magyar Számítógépes Nyelvészeti Konferencia}, year = {2022}, publisher = {Szegedi Tudományegyetem, Informatikai Intézet}, address = {Szeged, Magyarország}, author = {{Yang Zijian Győző}}, pages = {15--29} } ```
benjamin/gpt2-wechsel-chinese	feee72e42c2b685b2db8905223633ae6ce92f20f	2022-07-13T23:43:53.000Z	[ "pytorch", "gpt2", "text-generation", "zh", "transformers", "license:mit" ]	text-generation	false	benjamin	null	benjamin/gpt2-wechsel-chinese	86	null	transformers	4,874	--- language: zh license: mit --- # gpt2-wechsel-chinese Model trained with WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models. See the code here: https://github.com/CPJKU/wechsel And the paper here: https://aclanthology.org/2022.naacl-main.293/ ## Performance ### RoBERTa \| Model \| NLI Score \| NER Score \| Avg Score \| \|---\|---\|---\|---\| \| `roberta-base-wechsel-french` \| 82.43 \| 90.88 \| 86.65 \| \| `camembert-base` \| 80.88 \| 90.26 \| 85.57 \| \| Model \| NLI Score \| NER Score \| Avg Score \| \|---\|---\|---\|---\| \| `roberta-base-wechsel-german` \| 81.79 \| 89.72 \| 85.76 \| \| `deepset/gbert-base` \| 78.64 \| 89.46 \| 84.05 \| \| Model \| NLI Score \| NER Score \| Avg Score \| \|---\|---\|---\|---\| \| `roberta-base-wechsel-chinese` \| 78.32 \| 80.55 \| 79.44 \| \| `bert-base-chinese` \| 76.55 \| 82.05 \| 79.30 \| \| Model \| NLI Score \| NER Score \| Avg Score \| \|---\|---\|---\|---\| \| `roberta-base-wechsel-swahili` \| 75.05 \| 87.39 \| 81.22 \| \| `xlm-roberta-base` \| 69.18 \| 87.37 \| 78.28 \| ### GPT2 \| Model \| PPL \| \|---\|---\| \| `gpt2-wechsel-french` \| 19.71 \| \| `gpt2` (retrained from scratch) \| 20.47 \| \| Model \| PPL \| \|---\|---\| \| `gpt2-wechsel-german` \| 26.8 \| \| `gpt2` (retrained from scratch) \| 27.63 \| \| Model \| PPL \| \|---\|---\| \| `gpt2-wechsel-chinese` \| 51.97 \| \| `gpt2` (retrained from scratch) \| 52.98 \| \| Model \| PPL \| \|---\|---\| \| `gpt2-wechsel-swahili` \| 10.14 \| \| `gpt2` (retrained from scratch) \| 10.58 \| See our paper for details. ## Citation Please cite WECHSEL as ``` @inproceedings{minixhofer-etal-2022-wechsel, title = "{WECHSEL}: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models", author = "Minixhofer, Benjamin and Paischer, Fabian and Rekabsaz, Navid", booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies", month = jul, year = "2022", address = "Seattle, United States", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2022.naacl-main.293", pages = "3992--4006", abstract = "Large pretrained language models (LMs) have become the central building block of many NLP applications. Training these models requires ever more computational resources and most of the existing models are trained on English text only. It is exceedingly expensive to train these models in other languages. To alleviate this problem, we introduce a novel method {--} called WECHSEL {--} to efficiently and effectively transfer pretrained LMs to new languages. WECHSEL can be applied to any model which uses subword-based tokenization and learns an embedding for each subword. The tokenizer of the source model (in English) is replaced with a tokenizer in the target language and token embeddings are initialized such that they are semantically similar to the English tokens by utilizing multilingual static word embeddings covering English and the target language. We use WECHSEL to transfer the English RoBERTa and GPT-2 models to four languages (French, German, Chinese and Swahili). We also study the benefits of our method on very low-resource languages. WECHSEL improves over proposed methods for cross-lingual parameter transfer and outperforms models of comparable size trained from scratch with up to 64x less training effort. Our method makes training large language models for new languages more accessible and less damaging to the environment. We make our code and models publicly available.", } ```
nates-test-org/convit_small	0d0cc307c82f89aa6268ef668e056b1ba00c2fc1	2021-10-29T04:45:04.000Z	[ "pytorch", "timm", "image-classification" ]	image-classification	false	nates-test-org	null	nates-test-org/convit_small	86	null	timm	4,875	--- tags: - image-classification - timm library_tag: timm --- # Model card for convit_small
persiannlp/mt5-base-parsinlu-squad-reading-comprehension	626121fa4a6a18daa743231abe29b3419c03cd61	2021-09-23T16:20:07.000Z	[ "pytorch", "mt5", "text2text-generation", "fa", "multilingual", "dataset:parsinlu", "dataset:squad", "transformers", "reading-comprehension", "persian", "farsi", "license:cc-by-nc-sa-4.0", "autotrain_compatible" ]	text2text-generation	false	persiannlp	null	persiannlp/mt5-base-parsinlu-squad-reading-comprehension	86	1	transformers	4,876	--- language: - fa - multilingual thumbnail: https://upload.wikimedia.org/wikipedia/commons/a/a2/Farsi.svg tags: - reading-comprehension - mt5 - persian - farsi license: cc-by-nc-sa-4.0 datasets: - parsinlu - squad metrics: - f1 --- # Reading Comprehension (مدل برای پاسخ به درک مطلب) This is a mT5-based model for reading comprehension. Here is an example of how you can run this model: ```python from transformers import MT5ForConditionalGeneration, MT5Tokenizer model_size = "base" model_name = f"persiannlp/mt5-{model_size}-parsinlu-squad-reading-comprehension" tokenizer = MT5Tokenizer.from_pretrained(model_name) model = MT5ForConditionalGeneration.from_pretrained(model_name) def run_model(paragraph, question, generator_args): input_ids = tokenizer.encode(question + "\n" + paragraph, return_tensors="pt") res = model.generate(input_ids, generator_args) output = tokenizer.batch_decode(res, skip_special_tokens=True) print(output) return output run_model( "یک شی را دارای تقارن می‌نامیم زمانی که ان شی را بتوان به دو یا چند قسمت تقسیم کرد که آن‌ها قسمتی از یک طرح سازمان یافته باشند یعنی بر روی شکل تنها جابجایی و چرخش و بازتاب و تجانس انجام شود و در اصل شکل تغییری به وجود نیایید آنگاه ان را تقارن می‌نامیم مرکز تقارن:اگر در یک شکل نقطه‌ای مانندA وجود داشته باشد که هر نقطهٔ روی شکل (محیط) نسبت به نقطه یAمتقارن یک نقطهٔ دیگر شکل (محیط) باشد، نقطهٔ Aمرکز تقارن است. یعنی هر نقطه روی شکل باید متقارنی داشته باشد شکل‌های که منتظم هستند و زوج ضلع دارند دارای مرکز تقارند ولی شکل‌های فرد ضلعی منتظم مرکز تقارن ندارند. متوازی‌الأضلاع و دایره یک مرکز تقارن دارند ممکن است یک شکل خط تقارن نداشته باشد ولی مرکز تقارن داشته باشد. (منبع:س. گ)", "اشکالی که یک مرکز تقارن دارند" ) run_model( "شُتُر یا اُشتر را که در زبان پهلوی (ushtar)[نیازمند منبع] می‌گفتند حیوانی است نیرومند و تنومند با توش و توان بالا از خانواده شتران؛ شبه نشخوارکننده و با دست و گردنی دراز. بر پشت خود یک یا دو کوهان دارد که ساختارش از پیه و چربی است. در دین اسلام گوشت او حلال است. اما ذبح آن با دیگر جانوران حلال گوشت متفاوت است و آن را نحر (بریدن گلو) می‌کنند و اگر سر آن را مانند گوسفند پیش از نحر ببرند گوشت آن حلال نیست. شیرش نیز نوشیده می‌شود ولی بیشتر کاربرد بارکشی دارد. پشم و پوستش نیز برای ریسندگی و پارچه‌بافی و کفش‌دوزی کاربرد دارد. گونه‌های دیگری از شتران نیز در آمریکای جنوبی زندگی می‌کنند، به نام‌های لاما، آلپاکا، گواناکو که دارای کوهان نیستند. شتر ویژگی‌های خاصّی دارد که مهم‌ترین آن‌ها تحمّل شرایط سخت صحرا و دماهای گوناگون و به‌ویژه گرمای شدید تابستان و کمبود آب و علوفه است. ترکیب جسمانی شتر با دیگر جانوران اختلاف زیادی دارد، و این اختلاف انگیزه شده که شتر در درازا روزهای سال در بیابان زندگی کند و از بوته‌ها و درختچه‌های گوناگون صحرایی و کویری و حتی از بوته‌های شور و خاردار تغذیه کند. عرب‌ها از زمان‌های بسیار دور از شتر استفاده کرده و می‌کنند. آن‌ها به این حیوان اهلی لقب کشتی صحرا (به عربی: سفینةالصحراء) داده‌اند.", "غذای شترچیست؟" ) run_model( """حسین میرزایی می‌گوید مرحله اول پرداخت وام حمایتی کرونا به همگی خانوارهای یارانه‌بگیر متقاضی تکمیل شده است و حال چهار میلیون خانوار که به عنوان "اقشار خاص" و "آسیب‌پذیر" شناسایی شدند، می‌توانند برای یک میلیون تومان وام دیگر درخواست بدهند. آقای میرزایی گفته خانوارهای "آسیب‌پذیر" که شرایط گرفتن وام یک میلیونی اضافی را دارند با پیامک از این امکان مطلع شده‌اند. بنا به گزارش‌های رسمی با شیوع کرونا در ایران یک میلیون نفر بیکار شده‌اند و درآمد کارکنان مشاغل غیررسمی نیز ضربه قابل توجهی خورده است. ارزش ریال هم در هفته‌های اخیر در برابر ارزهای خارجی سقوط کرده است. اقتصاد ایران پیش از شیوع کرونا نیز با مشکلات مزمن رکود، تورم، تحریم و فساد روبرو بود.""", "وام یارانه به چه کسانی میدهند؟" ) run_model( "در ۲۲ ژوئن ۱۹۴۱ نیروهای محور در عملیات بارباروسا حمله سنگینی به اتحاد شوروی کرده و یکی از بزرگترین نبردهای زمینی تاریخ بشر را رقم زدند. همچنین جبهه شرقی باعث به دام افتادن نیروهای محور شد و بیش از همه ارتش آلمان نازی را درگیر جنگ فرسایشی کرد. در دسامبر ۱۹۴۱ ژاپن یک در عملیاتی ناگهانی با نام نبرد پرل هاربر به پایگاه دریایی ایالات متحده آمریکا حمله کرد. به دنبال این اتفاق آمریکا نیز بلافاصله علیه ژاپن اعلان جنگ کرد که با حمایت بریتانیا همراه شد. پس از آن متحدین (نیروهای محور در اروپا) نیز با اتحاد ژاپن علیه آمریکا اعلام جنگ کردند. دست‌آوردهای ژاپن در یورش به آمریکا باعث ایجاد این احساس در آسیا شد که آسیا از تسلط غرب خارج شده‌است از این رو بسیاری از ارتش‌های شکست خورده با آنها همراهی کردند.", "چرا امریکا وارد جنگ جهانی دوم شد؟" ) ``` For more details, visit this page: https://github.com/persiannlp/parsinlu/
projecte-aina/roberta-base-ca-cased-ner	aa44b621c767260cafa37e060f4f8a251f619936	2022-06-15T07:55:56.000Z	[ "pytorch", "roberta", "token-classification", "ca", "dataset:projecte-aina/ancora-ca-ner", "arxiv:1907.11692", "transformers", "catalan", "named entity recognition", "ner", "CaText", "Catalan Textual Corpus", "license:apache-2.0", "model-index", "autotrain_compatible" ]	token-classification	false	projecte-aina	null	projecte-aina/roberta-base-ca-cased-ner	86	1	transformers	4,877	--- language: - ca license: apache-2.0 tags: - "catalan" - "named entity recognition" - "ner" - "CaText" - "Catalan Textual Corpus" datasets: - "projecte-aina/ancora-ca-ner" metrics: - f1 model-index: - name: roberta-base-ca-cased-ner results: - task: type: token-classification dataset: type: projecte-aina/ancora-ca-ner name: ancora-ca-ner metrics: - type: f1 value: 0.8813 widget: - text: "Em dic Lluïsa i visc a Santa Maria del Camí." - text: "L'Aina, la Berta i la Norma són molt amigues." - text: "El Martí llegeix el Cavall Fort." --- # Catalan BERTa (RoBERTa-base) finetuned for Named Entity Recognition. The roberta-base-ca-cased-ner is a Named Entity Recognition (NER) model for the Catalan language fine-tuned from the [BERTa](https://huggingface.co/PlanTL-GOB-ES/roberta-base-ca) model, a [RoBERTa](https://arxiv.org/abs/1907.11692) base model pre-trained on a medium-size corpus collected from publicly available corpora and crawlers (check the BERTa model card for more details). ## Datasets We used the NER dataset in Catalan called [Ancora-ca-ner](https://huggingface.co/datasets/projecte-aina/ancora-ca-ner) for training and evaluation. ## Evaluation and results We evaluated the _roberta-base-ca-cased-ner_ on the Ancora-ca-ner test set against standard multilingual and monolingual baselines: \| Model \| Ancora-ca-ner (F1)\| \| ------------\|:-------------\| \| roberta-base-ca-cased-ner \| 88.13 \| \| mBERT \| 86.38 \| \| XLM-RoBERTa \| 87.66 \| \| WikiBERT-ca \| 77.66 \| For more details, check the fine-tuning and evaluation scripts in the official [GitHub repository](https://github.com/projecte-aina/club). ## Citing If you use any of these resources (datasets or models) in your work, please cite our latest paper: ```bibtex @inproceedings{armengol-estape-etal-2021-multilingual, title = "Are Multilingual Models the Best Choice for Moderately Under-resourced Languages? {A} Comprehensive Assessment for {C}atalan", author = "Armengol-Estap{\'e}, Jordi and Carrino, Casimiro Pio and Rodriguez-Penagos, Carlos and de Gibert Bonet, Ona and Armentano-Oller, Carme and Gonzalez-Agirre, Aitor and Melero, Maite and Villegas, Marta", booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021", month = aug, year = "2021", address = "Online", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2021.findings-acl.437", doi = "10.18653/v1/2021.findings-acl.437", pages = "4933--4946", } ```
remi/bertabs-finetuned-xsum-extractive-abstractive-summarization	a184879f346b3a77c35e9390c7e4a660cb2ef6e3	2021-05-20T04:17:40.000Z	[ "pytorch", "bert", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	remi	null	remi/bertabs-finetuned-xsum-extractive-abstractive-summarization	86	null	transformers	4,878	Entry not found
nickmuchi/vit-finetuned-chest-xray-pneumonia	2086fb41a4a93b6c7bf701ae8d068157591c8afb	2022-03-09T12:50:04.000Z	[ "pytorch", "tensorboard", "vit", "image-classification", "dataset:chest X-rays", "transformers", "generated_from_trainer", "license:apache-2.0", "model-index" ]	image-classification	false	nickmuchi	null	nickmuchi/vit-finetuned-chest-xray-pneumonia	86	null	transformers	4,879	--- license: apache-2.0 tags: - image-classification - generated_from_trainer metrics: - accuracy datasets: - chest X-rays widget: - src: https://drive.google.com/uc?id=1ygVCyEn6mfsNwpT1ZvWxANg5_DvStA7M example_title: PNEUMONIA - src: https://drive.google.com/uc?id=1xjcIEDb8kuSd4wF44gCEgsc0PfRvs53m example_title: NORMAL model-index: - name: vit-finetuned-chest-xray-pneumonia 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. --> # vit-finetuned-chest-xray-pneumonia This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the [chest-xray-pneumonia](https://www.kaggle.com/paultimothymooney/chest-xray-pneumonia) dataset. It achieves the following results on the evaluation set: - Loss: 0.1271 - Accuracy: 0.9551 ## 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: 16 - 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 \| Accuracy \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:--------:\| \| No log \| 1.0 \| 326 \| 0.2739 \| 0.9167 \| \| 0.2238 \| 2.0 \| 652 \| 0.2892 \| 0.9071 \| \| 0.2238 \| 3.0 \| 978 \| 0.2077 \| 0.9407 \| \| 0.1385 \| 4.0 \| 1304 \| 0.1349 \| 0.9535 \| \| 0.1347 \| 5.0 \| 1630 \| 0.1271 \| 0.9551 \| \| 0.1347 \| 6.0 \| 1956 \| 0.1458 \| 0.9535 \| \| 0.1112 \| 7.0 \| 2282 \| 0.2040 \| 0.9375 \| \| 0.1063 \| 8.0 \| 2608 \| 0.1423 \| 0.9567 \| \| 0.1063 \| 9.0 \| 2934 \| 0.1473 \| 0.9535 \| \| 0.0944 \| 10.0 \| 3260 \| 0.1385 \| 0.9583 \| ## Example Images #### Pneumonia Chest X-Ray ![Pneumonia](https://drive.google.com/uc?id=1yqnhD4Wjt4Y_NGLtijTGGaaw9GL497kQ) #### Normal Chest X-Ray ![Normal](https://drive.google.com/uc?id=1xjcIEDb8kuSd4wF44gCEgsc0PfRvs53m) ### Framework versions - Transformers 4.17.0 - Pytorch 1.10.0+cu111 - Datasets 1.18.4 - Tokenizers 0.11.6
Alvenir/wav2vec2-base-da-ft-nst	f8e0c5370b6db09ff54eb8a15ed642f58eaae55f	2022-03-17T16:16:12.000Z	[ "pytorch", "wav2vec2", "automatic-speech-recognition", "da", "transformers", "speech-to-text", "license:apache-2.0" ]	automatic-speech-recognition	false	Alvenir	null	Alvenir/wav2vec2-base-da-ft-nst	86	3	transformers	4,880	--- language: da tags: - speech-to-text license: apache-2.0 --- # wav2vec2-base-da-ft-nst This the [alvenir wav2vec2 model](https://huggingface.co/Alvenir/wav2vec2-base-da) for Danish ASR finetuned by Alvenir on the public NST dataset. The model is trained on 16kHz, so make sure your data is the same sample rate. The model was trained using fairseq and then converted to huggingface/transformers format. Alvenir is always happy to help with your own open-source ASR projects, customized domain specializations or premium models. ;-) ## Usage ```Python import soundfile as sf import torch from transformers import Wav2Vec2CTCTokenizer, Wav2Vec2Tokenizer, Wav2Vec2Processor, \ Wav2Vec2ForCTC def get_tokenizer(model_path: str) -> Wav2Vec2CTCTokenizer: return Wav2Vec2Tokenizer.from_pretrained(model_path) def get_processor(model_path: str) -> Wav2Vec2Processor: return Wav2Vec2Processor.from_pretrained(model_path) def load_model(model_path: str) -> Wav2Vec2ForCTC: return Wav2Vec2ForCTC.from_pretrained(model_path) model_id = "Alvenir/wav2vec2-base-da-ft-nst" model = load_model(model_id) model.eval() tokenizer = get_tokenizer(model_id) processor = get_processor(model_id) audio_file = "<path/to/audio.wav>" audio, _ = sf.read(audio_file) input_values = processor(audio, return_tensors="pt", padding="longest", sampling_rate=16_000).input_values with torch.no_grad(): logits = model(input_values).logits predicted_ids = torch.argmax(logits, dim=-1) transcription = processor.batch_decode(predicted_ids) print(transcription) ``` ## Benchmark results This is some benchmark results on the public available datasets in Danish. \| Dataset \| WER Greedy \| WER with 3-gram Language Model \| \|---------------------\|------------\|--------------------\| \| NST test \| 15,8% \| 11.9% \| \| alvenir-asr-da-eval \| 19.0% \| 12.1% \| \| common_voice_80 da test \| 26,3% \| 19,2% \|
ai4bharat/MultiIndicParaphraseGeneration	3f5c5a06fa624a6267d93df3e8332197cc5cb6f5	2022-03-31T06:21:30.000Z	[ "pytorch", "mbart", "text2text-generation", "as", "bn", "gu", "hi", "kn", "ml", "mr", "or", "pa", "ta", "te", "dataset:ai4bharat/IndicParaphrase", "arxiv:2203.05437", "transformers", "paraphrase-generation", "multilingual", "nlp", "indicnlp", "license:mit", "autotrain_compatible" ]	text2text-generation	false	ai4bharat	null	ai4bharat/MultiIndicParaphraseGeneration	86	null	transformers	4,881	--- tags: - paraphrase-generation - multilingual - nlp - indicnlp datasets: - ai4bharat/IndicParaphrase language: - as - bn - gu - hi - kn - ml - mr - or - pa - ta - te license: - mit --- # MultiIndicParaphraseGeneration This repository contains the [IndicBART](https://huggingface.co/ai4bharat/IndicBART) checkpoint finetuned on the 11 languages of [IndicParaphrase](https://huggingface.co/datasets/ai4bharat/IndicParaphrase) dataset. For finetuning details, see the [paper](https://arxiv.org/abs/2203.05437). <ul> <li >Supported languages: Assamese, Bengali, Gujarati, Hindi, Marathi, Odiya, Punjabi, Kannada, Malayalam, Tamil, and Telugu. Not all of these languages are supported by mBART50 and mT5. </li> <li >The model is much smaller than the mBART and mT5(-base) models, so less computationally expensive for decoding. </li> <li> Trained on large Indic language corpora (5.53 million sentences). </li> <li> All languages, have been represented in Devanagari script to encourage transfer learning among the related languages. </li> </ul> ## Using this model in `transformers` ``` from transformers import MBartForConditionalGeneration, AutoModelForSeq2SeqLM from transformers import AlbertTokenizer, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained("ai4bharat/MultiIndicParaphraseGeneration", do_lower_case=False, use_fast=False, keep_accents=True) # Or use tokenizer = AlbertTokenizer.from_pretrained("ai4bharat/MultiIndicParaphraseGeneration", do_lower_case=False, use_fast=False, keep_accents=True) model = AutoModelForSeq2SeqLM.from_pretrained("ai4bharat/MultiIndicParaphraseGeneration") # Or use model = MBartForConditionalGeneration.from_pretrained("ai4bharat/MultiIndicParaphraseGeneration") # Some initial mapping bos_id = tokenizer._convert_token_to_id_with_added_voc("<s>") eos_id = tokenizer._convert_token_to_id_with_added_voc("</s>") pad_id = tokenizer._convert_token_to_id_with_added_voc("<pad>") # To get lang_id use any of ['<2as>', '<2bn>', '<2en>', '<2gu>', '<2hi>', '<2kn>', '<2ml>', '<2mr>', '<2or>', '<2pa>', '<2ta>', '<2te>'] # First tokenize the input. The format below is how IndicBART was trained so the input should be "Sentence </s> <2xx>" where xx is the language code. Similarly, the output should be "<2yy> Sentence </s>". inp = tokenizer("दिल्ली यूनिवर्सिटी देश की प्रसिद्ध यूनिवर्सिटी में से एक है. </s> <2hi>", add_special_tokens=False, return_tensors="pt", padding=True).input_ids # For generation. Pardon the messiness. Note the decoder_start_token_id. model_output=model.generate(inp, use_cache=True,no_repeat_ngram_size=3,encoder_no_repeat_ngram_size=3, num_beams=4, max_length=20, min_length=1, early_stopping=True, pad_token_id=pad_id, bos_token_id=bos_id, eos_token_id=eos_id, decoder_start_token_id=tokenizer._convert_token_to_id_with_added_voc("<2hi>")) # Decode to get output strings decoded_output=tokenizer.decode(model_output[0], skip_special_tokens=True, clean_up_tokenization_spaces=False) print(decoded_output) # दिल्ली विश्वविद्यालय देश की प्रमुख विश्वविद्यालयों में शामिल है। # Note that if your output language is not Hindi or Marathi, you should convert its script from Devanagari to the desired language using the Indic NLP Library. ``` # Note: If you wish to use any language written in a non-Devanagari script, then you should first convert it to Devanagari using the <a href="https://github.com/anoopkunchukuttan/indic_nlp_library">Indic NLP Library</a>. After you get the output, you should convert it back into the original script. ## Benchmarks Scores on the `IndicParaphrase` test sets are as follows: Language \| BLEU / Self-BLEU / iBLEU ---------\|---------------------------- as \| 1.66 / 2.06 / 0.54 bn \| 11.57 / 1.69 / 7.59 gu \| 22.10 / 2.76 / 14.64 hi \| 27.29 / 2.87 / 18.24 kn \| 15.40 / 2.98 / 9.89 ml \| 10.57 / 1.70 / 6.89 mr \| 20.38 / 2.20 / 13.61 or \| 19.26 / 2.10 / 12.85 pa \| 14.87 / 1.35 / 10.00 ta \| 18.52 / 2.88 / 12.10 te \| 16.70 / 3.34 / 10.69 ## Citation If you use this model, please cite the following paper: ``` @inproceedings{Kumar2022IndicNLGSM, title={IndicNLG Suite: Multilingual Datasets for Diverse NLG Tasks in Indic Languages}, author={Aman Kumar and Himani Shrotriya and Prachi Sahu and Raj Dabre and Ratish Puduppully and Anoop Kunchukuttan and Amogh Mishra and Mitesh M. Khapra and Pratyush Kumar}, year={2022}, url = "https://arxiv.org/abs/2203.05437" } ```
efederici/sentence-BERTino	871f99951d3375f75a9cf7dc147cb1e0fac0170f	2022-05-03T13:14:23.000Z	[ "pytorch", "distilbert", "feature-extraction", "it", "sentence-transformers", "sentence-similarity", "transformers", "license:apache-2.0" ]	sentence-similarity	false	efederici	null	efederici/sentence-BERTino	86	1	sentence-transformers	4,882	--- pipeline_tag: sentence-similarity license: apache-2.0 language: - it tags: - sentence-transformers - feature-extraction - sentence-similarity - transformers --- # sentence-BERTino 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. It was trained on a dataset made from question/context pairs ([squad-it](https://github.com/crux82/squad-it)) and tags/news-article pairs (via scraping). ## 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 = ["Questo è un esempio di frase", "Questo è un ulteriore esempio"] model = SentenceTransformer('efederici/sentence-BERTino') 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 = ["Questo è un esempio di frase", "Questo è un ulteriore esempio"] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('efederici/sentence-BERTino') model = AutoModel.from_pretrained('efederici/sentence-BERTino') # 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) ``` ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DistilBertModel (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}) ) ```
NYTK/named-entity-recognition-nerkor-hubert-hungarian	a52a08bcc8abe9447ccc860f648b34500c130338	2022-04-01T09:03:27.000Z	[ "pytorch", "bert", "token-classification", "hu", "transformers", "license:gpl", "autotrain_compatible" ]	token-classification	false	NYTK	null	NYTK/named-entity-recognition-nerkor-hubert-hungarian	86	null	transformers	4,883	--- language: - hu tags: - token-classification license: gpl metrics: - f1 widget: - text: "A Kovácsné Nagy Erzsébet nagyon jól érzi magát a Nokiánál, azonban a Németországból érkezett Kovács Péter nehezen boldogul a beilleszkedéssel." --- # Hungarian Sentence-level Sentiment Analysis model with XLM-RoBERTa For further models, scripts and details, see [our demo site](https://juniper.nytud.hu/demo/nlp). - Pretrained model used: SZTAKI-HLT/hubert-base-cc - Finetuned on [NYTK-NerKor](https://github.com/nytud/NYTK-NerKor) - NE categories are: PER, LOC, MISC, ORG ## Limitations - max_seq_length = 128 ## Results F-score: 90.18% ## Citation If you use this model, please cite the following paper: ``` @inproceedings {yang-language-models, title = {Training language models with low resources: RoBERTa, BART and ELECTRA experimental models for Hungarian}, booktitle = {Proceedings of 12th IEEE International Conference on Cognitive Infocommunications (CogInfoCom 2021)}, year = {2021}, publisher = {IEEE}, address = {Online}, author = {{Yang, Zijian Győző and Váradi, Tamás}}, pages = {279--285} } ```
IDEA-CCNL/Erlangshen-Roberta-330M-NLI	f42d4a45e9dc3933398be40431315edbd4e19c21	2022-05-12T09:49:11.000Z	[ "pytorch", "bert", "text-classification", "zh", "transformers", "NLU", "NLI", "license:apache-2.0" ]	text-classification	false	IDEA-CCNL	null	IDEA-CCNL/Erlangshen-Roberta-330M-NLI	86	null	transformers	4,884	--- language: - zh license: apache-2.0 tags: - bert - NLU - NLI inference: true widget: - text: "今天心情不好[SEP]今天很开心" --- # Erlangshen-Roberta-330M-NLI, model (Chinese)，one model of [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM). We collect 4 NLI（Natural Language Inference） datasets in the Chinese domain for finetune, with a total of 1014787 samples. Our model is mainly based on [roberta](https://huggingface.co/hfl/chinese-roberta-wwm-ext-large) ## Usage ```python from transformers import BertForSequenceClassification from transformers import BertTokenizer import torch tokenizer=BertTokenizer.from_pretrained('IDEA-CCNL/Erlangshen-Roberta-330M-NLI') model=BertForSequenceClassification.from_pretrained('IDEA-CCNL/Erlangshen-Roberta-330M-NLI') texta='今天的饭不好吃' textb='今天心情不好' output=model(torch.tensor([tokenizer.encode(texta,textb)])) print(torch.nn.functional.softmax(output.logits,dim=-1)) ``` ## Scores on downstream chinese tasks (without any data augmentation) \| Model \| cmnli \| ocnli \| snli \| \| :--------: \| :-----: \| :----: \| :-----: \| \| Erlangshen-Roberta-110M-NLI \| 80.83 \| 78.56 \| 88.01 \| \| Erlangshen-Roberta-330M-NLI \| 82.25 \| 79.82 \| 88 \| \| Erlangshen-MegatronBert-1.3B-NLI \| 84.52 \| 84.17 \| 88.67 \| ## Citation If you find the resource is useful, please cite the following website in your paper. ``` @misc{Fengshenbang-LM, title={Fengshenbang-LM}, author={IDEA-CCNL}, year={2021}, howpublished={\url{https://github.com/IDEA-CCNL/Fengshenbang-LM}}, } ```
mary905el/ruT5_neuro_chgk_answering	349381b078e48288d6b82e26d5cd5c68c5c369bb	2022-04-27T05:33:03.000Z	[ "pytorch", "t5", "text2text-generation", "ru", "transformers", "PyTorch", "Transformers", "autotrain_compatible" ]	text2text-generation	false	mary905el	null	mary905el/ruT5_neuro_chgk_answering	86	null	transformers	4,885	--- language: - ru tags: - PyTorch - Transformers widget: - text: "Ответьте двумя словами, что мы заменили на ИКС?" inference: parameters: do_sample: True temperature: 0.8 --- This is https://huggingface.co/sberbank-ai/ruT5-base model, fine-tuned to answer ChGK questions (Что? Где? Когда? https://db.chgk.info/) Dataset: 75 000 questions from 2000-2019 Trained for 10 epochs
smc/Electric_Pole_with_or_without_transformer	dc05ec14a729ac37195e2d0361cf279727d5f862	2022-05-21T22:09:33.000Z	[ "pytorch", "tensorboard", "vit", "image-classification", "transformers", "huggingpics", "model-index" ]	image-classification	false	smc	null	smc/Electric_Pole_with_or_without_transformer	86	1	transformers	4,886	--- tags: - image-classification - pytorch - huggingpics metrics: - accuracy model-index: - name: Electric_Pole_with_or_without_transformer results: - task: name: Image Classification type: image-classification metrics: - name: Accuracy type: accuracy value: 0.9230769276618958 --- # Electric_Pole_with_or_without_transformer Classify your electric pole images
qalover/chinese-pert-large-open-domain-mrc	e9cfdc887fe9d6cb043fb9b870e1908067e31916	2022-05-31T14:21:54.000Z	[ "pytorch", "bert", "question-answering", "zh", "transformers", "license:gpl-3.0", "autotrain_compatible" ]	question-answering	false	qalover	null	qalover/chinese-pert-large-open-domain-mrc	86	1	transformers	4,887	--- language: - zh license: gpl-3.0 --- ## 基于 chinese-pert-large 训练的面向开放领域MRC 模型使用中文MRC数据（cmrc2018, webqa与laisi的训练集）训练的chinese-pert-large模型 ## 训练过程使用了[UER-py](https://github.com/dbiir/UER-py/) 进行fine-tuned 加入了包括但不限于摘要、负采样、混淆等数据加强方法并转换为Huggingface进行上传 \| \| CMRC 2018 Dev \| DRCD Dev \| SQuAD-Zen Dev (Answerable) \| AVG \| \| :-------: \| :-----------: \| :-------: \| :------------------------: \| :-------: \| \| PERT-large \| 74.4/89.8 \| 90.3/94.\| 62.8/78.8 \| 75.9/87.8 \|
renjithks/layoutlmv2-er-ner	2edce8c45d8a8307630be7e53eff3d362a74bb6e	2022-06-08T19:37:51.000Z	[ "pytorch", "tensorboard", "layoutlmv2", "token-classification", "transformers", "generated_from_trainer", "license:cc-by-nc-sa-4.0", "model-index", "autotrain_compatible" ]	token-classification	false	renjithks	null	renjithks/layoutlmv2-er-ner	86	null	transformers	4,888	--- license: cc-by-nc-sa-4.0 tags: - generated_from_trainer metrics: - precision - recall - f1 - accuracy model-index: - name: layoutlmv2-er-ner 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. --> # layoutlmv2-er-ner This model is a fine-tuned version of [renjithks/layoutlmv2-cord-ner](https://huggingface.co/renjithks/layoutlmv2-cord-ner) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.1217 - Precision: 0.7810 - Recall: 0.8085 - F1: 0.7945 - Accuracy: 0.9747 ## 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: 20 ### Training results \| Training Loss \| Epoch \| Step \| Validation Loss \| Precision \| Recall \| F1 \| Accuracy \| \|:-------------:\|:-----:\|:----:\|:---------------:\|:---------:\|:------:\|:------:\|:--------:\| \| No log \| 1.0 \| 41 \| 0.5441 \| 0.0 \| 0.0 \| 0.0 \| 0.8851 \| \| No log \| 2.0 \| 82 \| 0.4660 \| 0.1019 \| 0.0732 \| 0.0852 \| 0.8690 \| \| No log \| 3.0 \| 123 \| 0.2506 \| 0.4404 \| 0.4828 \| 0.4606 \| 0.9240 \| \| No log \| 4.0 \| 164 \| 0.1725 \| 0.6120 \| 0.6076 \| 0.6098 \| 0.9529 \| \| No log \| 5.0 \| 205 \| 0.1387 \| 0.7204 \| 0.7245 \| 0.7225 \| 0.9671 \| \| No log \| 6.0 \| 246 \| 0.1237 \| 0.7742 \| 0.7747 \| 0.7745 \| 0.9722 \| \| No log \| 7.0 \| 287 \| 0.1231 \| 0.7619 \| 0.7554 \| 0.7586 \| 0.9697 \| \| No log \| 8.0 \| 328 \| 0.1199 \| 0.7994 \| 0.7719 \| 0.7854 \| 0.9738 \| \| No log \| 9.0 \| 369 \| 0.1197 \| 0.7937 \| 0.8113 \| 0.8024 \| 0.9741 \| \| No log \| 10.0 \| 410 \| 0.1284 \| 0.7581 \| 0.7597 \| 0.7589 \| 0.9690 \| \| No log \| 11.0 \| 451 \| 0.1172 \| 0.7792 \| 0.7848 \| 0.7820 \| 0.9738 \| \| No log \| 12.0 \| 492 \| 0.1192 \| 0.7913 \| 0.7970 \| 0.7941 \| 0.9743 \| \| 0.1858 \| 13.0 \| 533 \| 0.1175 \| 0.7960 \| 0.8006 \| 0.7983 \| 0.9753 \| \| 0.1858 \| 14.0 \| 574 \| 0.1184 \| 0.7724 \| 0.8034 \| 0.7876 \| 0.9740 \| \| 0.1858 \| 15.0 \| 615 \| 0.1171 \| 0.7882 \| 0.8142 \| 0.8010 \| 0.9756 \| \| 0.1858 \| 16.0 \| 656 \| 0.1195 \| 0.7829 \| 0.8070 \| 0.7948 \| 0.9745 \| \| 0.1858 \| 17.0 \| 697 \| 0.1209 \| 0.7810 \| 0.8006 \| 0.7906 \| 0.9743 \| \| 0.1858 \| 18.0 \| 738 \| 0.1241 \| 0.7806 \| 0.7963 \| 0.7884 \| 0.9740 \| \| 0.1858 \| 19.0 \| 779 \| 0.1222 \| 0.7755 \| 0.8027 \| 0.7889 \| 0.9742 \| \| 0.1858 \| 20.0 \| 820 \| 0.1217 \| 0.7810 \| 0.8085 \| 0.7945 \| 0.9747 \| ### Framework versions - Transformers 4.16.2 - Pytorch 1.9.0+cu111 - Datasets 1.18.4 - Tokenizers 0.11.6
fusing/latent-diffusion-text2im-large	030d95399e64c1ff7be798e62b9bc6ef6ec7bf3b	2022-07-18T13:31:27.000Z	[ "pytorch", "ldmbert", "arxiv:2112.10752", "transformers", "diffusion", "license:mit" ]	null	false	fusing	null	fusing/latent-diffusion-text2im-large	86	2	transformers	4,889	--- tags: - diffusion license: mit --- Latent Diffusion Paper: [High-Resolution Image Synthesis with Latent Diffusion Models](https://arxiv.org/abs/2112.10752) Abstract: By decomposing the image formation process into a sequential application of denoising autoencoders, diffusion models (DMs) achieve state-of-the-art synthesis results on image data and beyond. Additionally, their formulation allows for a guiding mechanism to control the image generation process without retraining. However, since these models typically operate directly in pixel space, optimization of powerful DMs often consumes hundreds of GPU days and inference is expensive due to sequential evaluations. To enable DM training on limited computational resources while retaining their quality and flexibility, we apply them in the latent space of powerful pretrained autoencoders. In contrast to previous work, training diffusion models on such a representation allows for the first time to reach a near-optimal point between complexity reduction and detail preservation, greatly boosting visual fidelity. By introducing cross-attention layers into the model architecture, we turn diffusion models into powerful and flexible generators for general conditioning inputs such as text or bounding boxes and high-resolution synthesis becomes possible in a convolutional manner. Our latent diffusion models (LDMs) achieve a new state of the art for image inpainting and highly competitive performance on various tasks, including unconditional image generation, semantic scene synthesis, and super-resolution, while significantly reducing computational requirements compared to pixel-based DMs. Code is available at this https URL. ## Usage ```python from diffusers import DiffusionPipeline ldm = DiffusionPipeline.from_pretrained("fusing/latent-diffusion-text2im-large") generator = torch.manual_seed(42) prompt = "A painting of a squirrel eating a burger" image = ldm([prompt], generator=generator, eta=0.3, guidance_scale=6.0, num_inference_steps=50) image_processed = image.cpu().permute(0, 2, 3, 1) image_processed = image_processed * 255. image_processed = image_processed.numpy().astype(np.uint8) image_pil = PIL.Image.fromarray(image_processed[0]) # save image image_pil.save("test.png") ``` ## Samples 1. "A street sign that reads Huggingface." ![sample_1](https://huggingface.co/datasets/valhalla/images/resolve/main/ldm-1.png) 2."A painting of a squirrel eating a burger" ![sample_2](https://huggingface.co/datasets/valhalla/images/resolve/main/ldm-2.png)
tornqvistmax/7cats_finetuned	f747e28016a4aacf52435f263855c669bcd421d1	2022-06-16T14:43:45.000Z	[ "pytorch", "xlm-roberta", "text-classification", "transformers" ]	text-classification	false	tornqvistmax	null	tornqvistmax/7cats_finetuned	86	null	transformers	4,890	Entry not found
Vlasta/DNADebertaK7	ba8d72bc65ce662e14e0ecf36a815ed003beb5f5	2022-07-05T23:37:40.000Z	[ "pytorch", "deberta", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	Vlasta	null	Vlasta/DNADebertaK7	86	null	transformers	4,891	Entry not found
Chrode/bert_prot_temp_classifier	e86766a768e54af94880b3955b93b6bc08bd624c	2022-07-04T07:54:58.000Z	[ "pytorch", "BertTempProtClassifier", "transformers" ]	null	false	Chrode	null	Chrode/bert_prot_temp_classifier	86	null	transformers	4,892	Entry not found
darragh/swinunetr-btcv-tiny	ce084bb73ca3af65aa33b79a8d00eab93550122e	2022-07-15T21:01:18.000Z	[ "pytorch", "en", "dataset:BTCV", "transformers", "btcv", "medical", "swin", "license:apache-2.0" ]	null	false	darragh	null	darragh/swinunetr-btcv-tiny	86	null	transformers	4,893	--- language: en tags: - btcv - medical - swin license: apache-2.0 datasets: - BTCV --- # Model Overview This repository contains the code for Swin UNETR [1,2]. Swin UNETR is the state-of-the-art on Medical Segmentation Decathlon (MSD) and Beyond the Cranial Vault (BTCV) Segmentation Challenge dataset. In [1], a novel methodology is devised for pre-training Swin UNETR backbone in a self-supervised manner. We provide the option for training Swin UNETR by fine-tuning from pre-trained self-supervised weights or from scratch. The source repository for the training of these models can be found [here](https://github.com/Project-MONAI/research-contributions/tree/main/SwinUNETR/BTCV). # Installing Dependencies Dependencies for training and inference can be installed using the model requirements : ``` bash pip install -r requirements.txt ``` # Intended uses & limitations You can use the raw model for dicom segmentation, but it's mostly intended to be fine-tuned on a downstream task. Note that this model is primarily aimed at being fine-tuned on tasks which segment CAT scans or MRIs on images in dicom format. Dicom meta data mostly differs across medical facilities, so if applying to a new dataset, the model should be finetuned. # How to use To install necessary dependencies, run the below in bash. ``` git clone https://github.com/darraghdog/Project-MONAI-research-contributions pmrc pip install -r pmrc/requirements.txt cd pmrc/SwinUNETR/BTCV ``` To load the model from the hub. ``` >>> from swinunetr import SwinUnetrModelForInference >>> model = SwinUnetrModelForInference.from_pretrained('darragh/swinunetr-btcv-tiny') ``` # Limitations and bias The training data used for this model is specific to CAT scans from certain health facilities and machines. Data from other facilities may difffer in image distributions, and may require finetuning of the models for best performance. # Evaluation results We provide several pre-trained models on BTCV dataset in the following. <table> <tr> <th>Name</th> <th>Dice (overlap=0.7)</th> <th>Dice (overlap=0.5)</th> <th>Feature Size</th> <th># params (M)</th> <th>Self-Supervised Pre-trained </th> </tr> <tr> <td>Swin UNETR/Base</td> <td>82.25</td> <td>81.86</td> <td>48</td> <td>62.1</td> <td>Yes</td> </tr> <tr> <td>Swin UNETR/Small</td> <td>79.79</td> <td>79.34</td> <td>24</td> <td>15.7</td> <td>No</td> </tr> <tr> <td>Swin UNETR/Tiny</td> <td>72.05</td> <td>70.35</td> <td>12</td> <td>4.0</td> <td>No</td> </tr> </table> # Data Preparation ![image](https://lh3.googleusercontent.com/pw/AM-JKLX0svvlMdcrchGAgiWWNkg40lgXYjSHsAAuRc5Frakmz2pWzSzf87JQCRgYpqFR0qAjJWPzMQLc_mmvzNjfF9QWl_1OHZ8j4c9qrbR6zQaDJWaCLArRFh0uPvk97qAa11HtYbD6HpJ-wwTCUsaPcYvM=w1724-h522-no?authuser=0) The training data is from the [BTCV challenge dataset](https://www.synapse.org/#!Synapse:syn3193805/wiki/217752). - Target: 13 abdominal organs including 1. Spleen 2. Right Kidney 3. Left Kideny 4.Gallbladder 5.Esophagus 6. Liver 7. Stomach 8.Aorta 9. IVC 10. Portal and Splenic Veins 11. Pancreas 12.Right adrenal gland 13.Left adrenal gland. - Task: Segmentation - Modality: CT - Size: 30 3D volumes (24 Training + 6 Testing) # Training See the source repository [here](https://github.com/Project-MONAI/research-contributions/tree/main/SwinUNETR/BTCV) for information on training. # BibTeX entry and citation info If you find this repository useful, please consider citing the following papers: ``` @inproceedings{tang2022self, title={Self-supervised pre-training of swin transformers for 3d medical image analysis}, author={Tang, Yucheng and Yang, Dong and Li, Wenqi and Roth, Holger R and Landman, Bennett and Xu, Daguang and Nath, Vishwesh and Hatamizadeh, Ali}, booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition}, pages={20730--20740}, year={2022} } @article{hatamizadeh2022swin, title={Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images}, author={Hatamizadeh, Ali and Nath, Vishwesh and Tang, Yucheng and Yang, Dong and Roth, Holger and Xu, Daguang}, journal={arXiv preprint arXiv:2201.01266}, year={2022} } ``` # References [1]: Tang, Y., Yang, D., Li, W., Roth, H.R., Landman, B., Xu, D., Nath, V. and Hatamizadeh, A., 2022. Self-supervised pre-training of swin transformers for 3d medical image analysis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 20730-20740). [2]: Hatamizadeh, A., Nath, V., Tang, Y., Yang, D., Roth, H. and Xu, D., 2022. Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images. arXiv preprint arXiv:2201.01266.
Helsinki-NLP/opus-mt-fr-vi	d7a313fa61fa59ee759fd877759de83fd244dce8	2021-01-18T08:49:25.000Z	[ "pytorch", "marian", "text2text-generation", "fr", "vi", "transformers", "translation", "license:apache-2.0", "autotrain_compatible" ]	translation	false	Helsinki-NLP	null	Helsinki-NLP/opus-mt-fr-vi	85	null	transformers	4,894	--- language: - fr - vi tags: - translation license: apache-2.0 --- ### fra-vie * source group: French * target group: Vietnamese * OPUS readme: [fra-vie](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/fra-vie/README.md) * model: transformer-align * source language(s): fra * target language(s): vie * model: transformer-align * pre-processing: normalization + SentencePiece (spm32k,spm32k) * download original weights: [opus-2020-06-17.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/fra-vie/opus-2020-06-17.zip) * test set translations: [opus-2020-06-17.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/fra-vie/opus-2020-06-17.test.txt) * test set scores: [opus-2020-06-17.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/fra-vie/opus-2020-06-17.eval.txt) ## Benchmarks \| testset \| BLEU \| chr-F \| \|-----------------------\|-------\|-------\| \| Tatoeba-test.fra.vie \| 31.1 \| 0.486 \| ### System Info: - hf_name: fra-vie - source_languages: fra - target_languages: vie - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/fra-vie/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['fr', 'vi'] - src_constituents: {'fra'} - tgt_constituents: {'vie', 'vie_Hani'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/fra-vie/opus-2020-06-17.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/fra-vie/opus-2020-06-17.test.txt - src_alpha3: fra - tgt_alpha3: vie - short_pair: fr-vi - chrF2_score: 0.486 - bleu: 31.1 - brevity_penalty: 0.985 - ref_len: 13219.0 - src_name: French - tgt_name: Vietnamese - train_date: 2020-06-17 - src_alpha2: fr - tgt_alpha2: vi - prefer_old: False - long_pair: fra-vie - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41
KrishParikh/gpt2_imdb_movie_plots	ab4a551b10ca184133e3c2a3e213b71d3495f65c	2021-11-21T20:11:06.000Z	[ "pytorch", "gpt2", "text-generation", "transformers", "generated_from_trainer", "model-index" ]	text-generation	false	KrishParikh	null	KrishParikh/gpt2_imdb_movie_plots	85	null	transformers	4,895	--- tags: - generated_from_trainer model-index: - name: gpt2-plot 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. --> # gpt2-plot This model is a fine-tuned version of [gpt2-medium](https://huggingface.co/gpt2-medium) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.8856 ## 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: 1 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5.0 - mixed_precision_training: Native AMP ### Training results ### Framework versions - Transformers 4.13.0.dev0 - Pytorch 1.9.0 - Datasets 1.15.1 - Tokenizers 0.10.3
SkolkovoInstitute/roberta_toxicity_classifier_v1	0aeddbf4acc227b80ca3a5e2409c26b0988639a2	2021-11-02T18:36:13.000Z	[ "pytorch", "roberta", "text-classification", "arxiv:1911.00536", "transformers" ]	text-classification	false	SkolkovoInstitute	null	SkolkovoInstitute/roberta_toxicity_classifier_v1	85	null	transformers	4,896	This model is a clone of [SkolkovoInstitute/roberta_toxicity_classifier](https://huggingface.co/SkolkovoInstitute/roberta_toxicity_classifier) trained on a disjoint dataset. While `roberta_toxicity_classifier` is used for evaluation of detoxification algorithms, `roberta_toxicity_classifier_v1` can be used within these algorithms, as in the paper [Text Detoxification using Large Pre-trained Neural Models](https://arxiv.org/abs/1911.00536).
TODBERT/TOD-DistilBERT-JNT-V1	c21cde3992721781e96604a7030de7bff81dc663	2020-08-26T18:39:56.000Z	[ "pytorch", "distilbert", "fill-mask", "transformers", "autotrain_compatible" ]	fill-mask	false	TODBERT	null	TODBERT/TOD-DistilBERT-JNT-V1	85	null	transformers	4,897	Entry not found
YituTech/conv-bert-medium-small	4889125682c36c29e56dd0a70717c8706ef1333a	2021-02-24T11:24:27.000Z	[ "pytorch", "tf", "convbert", "feature-extraction", "transformers" ]	feature-extraction	false	YituTech	null	YituTech/conv-bert-medium-small	85	null	transformers	4,898	Entry not found
dbernsohn/roberta-python	cb9ceb18059e5edec8431480d2a21f749b5b4fca	2021-05-20T15:57:13.000Z	[ "pytorch", "jax", "roberta", "fill-mask", "python", "dataset:code_search_net", "arxiv:1907.11692", "transformers", "autotrain_compatible" ]	fill-mask	false	dbernsohn	null	dbernsohn/roberta-python	85	3	transformers	4,899	# roberta-python --- language: python datasets: - code_search_net --- This is a [roberta](https://arxiv.org/pdf/1907.11692.pdf) pre-trained version on the [CodeSearchNet dataset](https://github.com/github/CodeSearchNet) for Python Mask Language Model mission. To load the model: (necessary packages: !pip install transformers sentencepiece) ```python from transformers import AutoTokenizer, AutoModelWithLMHead, pipeline tokenizer = AutoTokenizer.from_pretrained("dbernsohn/roberta-python") model = AutoModelWithLMHead.from_pretrained("dbernsohn/roberta-python") fill_mask = pipeline( "fill-mask", model=model, tokenizer=tokenizer ) ``` You can then use this model to fill masked words in a Python code. ```python code = """ new_dict = {} for k, v in my_dict.<mask>(): new_dict[k] = v**2 """.lstrip() pred = {x["token_str"].replace("Ġ", ""): x["score"] for x in fill_mask(code)} sorted(pred.items(), key=lambda kv: kv[1], reverse=True) # [('items', 0.7376779913902283), # ('keys', 0.16238391399383545), # ('values', 0.03965481370687485), # ('iteritems', 0.03346433863043785), # ('splitlines', 0.0032723243348300457)] ``` The whole training process and hyperparameters are in my [GitHub repo](https://github.com/DorBernsohn/CodeLM/tree/main/CodeMLM) > Created by [Dor Bernsohn](https://www.linkedin.com/in/dor-bernsohn-70b2b1146/)

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.