Add new SentenceTransformer model

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+  "word_embedding_dimension": 1024,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": false,
+  "pooling_mode_lasttoken": false,
+  "include_prompt": true
+}

README.md

@@ -0,0 +1,528 @@
+---
+base_model: jinaai/jina-embeddings-v3
+library_name: sentence-transformers
+metrics:
+- pearson_cosine
+- spearman_cosine
+pipeline_tag: sentence-similarity
+tags:
+- sentence-transformers
+- sentence-similarity
+- feature-extraction
+- generated_from_trainer
+- dataset_size:63802
+- loss:CoSENTLoss
+widget:
+- source_sentence: машинка детская самоходная бибикар желтый
+  sentences:
+  - 'машинка детская красная бибикар '
+  - моторное масло alpine dx1 5w 30 5л 0101662
+  - 'спинбайк schwinn ic7 '
+- source_sentence: 'велосипед stels saber 20 фиолетовый '
+  sentences:
+  - 'детские спортивные комплексы '
+  - 'велосипед bmx stels saber 20 v010 2020 '
+  - 50218 кабель ugreen hd132 hdmi zinc alloy optical fiber cable черный 40m
+- source_sentence: гидравличесские прессы
+  sentences:
+  - пресс гидравлический ручной механизмом
+  - ракетка для настольного тенниса fora 7
+  - 'объектив panasonic 20mm f1 7 asph ii h h020ae k '
+- source_sentence: 'бокс пластиковый монтажной платой щмп п 300х200х130 мм ip65 proxima
+    ящики щитки шкафы '
+  sentences:
+  - батарейный отсек для 4xаа открытый проволочные выводы разъем dcx2 1 battery holder
+    4xaa 6v dc
+  - 'bugera bc15 '
+  - 'бокс пластиковый монтажной платой щмп п 500х350х190 мм ip65 proxima ящики щитки
+    шкафы '
+- source_sentence: 'honor watch gs pro black '
+  sentences:
+  - 'honor watch gs pro white '
+  - трансформер pituso carlo hb gy 06 lemon
+  - 'электровелосипед колхозник volten greenline 500w '
+model-index:
+- name: SentenceTransformer based on jinaai/jina-embeddings-v3
+  results:
+  - task:
+      type: semantic-similarity
+      name: Semantic Similarity
+    dataset:
+      name: example dev
+      type: example-dev
+    metrics:
+    - type: pearson_cosine
+      value: 0.47736782328677585
+      name: Pearson Cosine
+    - type: spearman_cosine
+      value: 0.49693031448879005
+      name: Spearman Cosine
+---
+
+# SentenceTransformer based on jinaai/jina-embeddings-v3
+
+This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [jinaai/jina-embeddings-v3](https://huggingface.co/jinaai/jina-embeddings-v3). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
+
+## Model Details
+
+### Model Description
+- **Model Type:** Sentence Transformer
+- **Base model:** [jinaai/jina-embeddings-v3](https://huggingface.co/jinaai/jina-embeddings-v3) <!-- at revision 30996fea06f69ecd8382ee4f11e29acaf6b5405e -->
+- **Maximum Sequence Length:** 8194 tokens
+- **Output Dimensionality:** 1024 dimensions
+- **Similarity Function:** Cosine Similarity
+<!-- - **Training Dataset:** Unknown -->
+<!-- - **Language:** Unknown -->
+<!-- - **License:** Unknown -->
+
+### Model Sources
+
+- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
+- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
+- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
+
+### Full Model Architecture
+
+```
+SentenceTransformer(
+  (transformer): Transformer(
+    (auto_model): XLMRobertaLoRA(
+      (roberta): XLMRobertaModel(
+        (embeddings): XLMRobertaEmbeddings(
+          (word_embeddings): ParametrizedEmbedding(
+            250002, 1024, padding_idx=1
+            (parametrizations): ModuleDict(
+              (weight): ParametrizationList(
+                (0): LoRAParametrization()
+              )
+            )
+          )
+          (token_type_embeddings): ParametrizedEmbedding(
+            1, 1024
+            (parametrizations): ModuleDict(
+              (weight): ParametrizationList(
+                (0): LoRAParametrization()
+              )
+            )
+          )
+        )
+        (emb_drop): Dropout(p=0.1, inplace=False)
+        (emb_ln): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
+        (encoder): XLMRobertaEncoder(
+          (layers): ModuleList(
+            (0-23): 24 x Block(
+              (mixer): MHA(
+                (rotary_emb): RotaryEmbedding()
+                (Wqkv): ParametrizedLinearResidual(
+                  in_features=1024, out_features=3072, bias=True
+                  (parametrizations): ModuleDict(
+                    (weight): ParametrizationList(
+                      (0): LoRAParametrization()
+                    )
+                  )
+                )
+                (inner_attn): SelfAttention(
+                  (drop): Dropout(p=0.1, inplace=False)
+                )
+                (inner_cross_attn): CrossAttention(
+                  (drop): Dropout(p=0.1, inplace=False)
+                )
+                (out_proj): ParametrizedLinear(
+                  in_features=1024, out_features=1024, bias=True
+                  (parametrizations): ModuleDict(
+                    (weight): ParametrizationList(
+                      (0): LoRAParametrization()
+                    )
+                  )
+                )
+              )
+              (dropout1): Dropout(p=0.1, inplace=False)
+              (drop_path1): StochasticDepth(p=0.0, mode=row)
+              (norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
+              (mlp): Mlp(
+                (fc1): ParametrizedLinear(
+                  in_features=1024, out_features=4096, bias=True
+                  (parametrizations): ModuleDict(
+                    (weight): ParametrizationList(
+                      (0): LoRAParametrization()
+                    )
+                  )
+                )
+                (fc2): ParametrizedLinear(
+                  in_features=4096, out_features=1024, bias=True
+                  (parametrizations): ModuleDict(
+                    (weight): ParametrizationList(
+                      (0): LoRAParametrization()
+                    )
+                  )
+                )
+              )
+              (dropout2): Dropout(p=0.1, inplace=False)
+              (drop_path2): StochasticDepth(p=0.0, mode=row)
+              (norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
+            )
+          )
+        )
+        (pooler): XLMRobertaPooler(
+          (dense): ParametrizedLinear(
+            in_features=1024, out_features=1024, bias=True
+            (parametrizations): ModuleDict(
+              (weight): ParametrizationList(
+                (0): LoRAParametrization()
+              )
+            )
+          )
+          (activation): Tanh()
+        )
+      )
+    )
+  )
+  (pooler): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
+  (normalizer): Normalize()
+)
+```
+
+## Usage
+
+### Direct Usage (Sentence Transformers)
+
+First install the Sentence Transformers library:
+
+```bash
+pip install -U sentence-transformers
+```
+
+Then you can load this model and run inference.
+```python
+from sentence_transformers import SentenceTransformer
+
+# Download from the 🤗 Hub
+model = SentenceTransformer("seregadgl/t2")
+# Run inference
+sentences = [
+    'honor watch gs pro black ',
+    'honor watch gs pro white ',
+    'трансформер pituso carlo hb gy 06 lemon',
+]
+embeddings = model.encode(sentences)
+print(embeddings.shape)
+# [3, 1024]
+
+# Get the similarity scores for the embeddings
+similarities = model.similarity(embeddings, embeddings)
+print(similarities.shape)
+# [3, 3]
+```
+
+<!--
+### Direct Usage (Transformers)
+
+<details><summary>Click to see the direct usage in Transformers</summary>
+
+</details>
+-->
+
+<!--
+### Downstream Usage (Sentence Transformers)
+
+You can finetune this model on your own dataset.
+
+<details><summary>Click to expand</summary>
+
+</details>
+-->
+
+<!--
+### Out-of-Scope Use
+
+*List how the model may foreseeably be misused and address what users ought not to do with the model.*
+-->
+
+## Evaluation
+
+### Metrics
+
+#### Semantic Similarity
+
+* Dataset: `example-dev`
+* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
+
+| Metric              | Value      |
+|:--------------------|:-----------|
+| pearson_cosine      | 0.4774     |
+| **spearman_cosine** | **0.4969** |
+
+<!--
+## Bias, Risks and Limitations
+
+*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
+-->
+
+<!--
+### Recommendations
+
+*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
+-->
+
+## Training Details
+
+### Training Dataset
+
+#### Unnamed Dataset
+
+
+* Size: 63,802 training samples
+* Columns: <code>doc</code>, <code>candidate</code>, and <code>label</code>
+* Approximate statistics based on the first 1000 samples:
+  |         | doc                                                                               | candidate                                                                         | label                                           |
+  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------|
+  | type    | string                                                                            | string                                                                            | int                                             |
+  | details | <ul><li>min: 3 tokens</li><li>mean: 14.82 tokens</li><li>max: 55 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 14.58 tokens</li><li>max: 37 tokens</li></ul> | <ul><li>0: ~85.20%</li><li>1: ~14.80%</li></ul> |
+* Samples:
+  | doc                                                    | candidate                                                              | label          |
+  |:-------------------------------------------------------|:-----------------------------------------------------------------------|:---------------|
+  | <code>массажер xiaomi massage gun eu bhr5608eu </code> | <code>перкуссионный массажер xiaomi massage gun mini bhr6083gl </code> | <code>0</code> |
+  | <code>безударная дрель ingco ed50028 </code>           | <code>ударная дрель ingco id211002 </code>                             | <code>0</code> |
+  | <code>жидкость old smuggler 30мл 20мг </code>          | <code>жидкость old smuggler salt 30ml marlboro 20mg</code>             | <code>0</code> |
+* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
+  ```json
+  {
+      "scale": 20.0,
+      "similarity_fct": "pairwise_cos_sim"
+  }
+  ```
+
+### Evaluation Dataset
+
+#### Unnamed Dataset
+
+
+* Size: 7,090 evaluation samples
+* Columns: <code>doc</code>, <code>candidate</code>, and <code>label</code>
+* Approximate statistics based on the first 1000 samples:
+  |         | doc                                                                               | candidate                                                                         | label                                           |
+  |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------|
+  | type    | string                                                                            | string                                                                            | int                                             |
+  | details | <ul><li>min: 4 tokens</li><li>mean: 14.91 tokens</li><li>max: 72 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 14.56 tokens</li><li>max: 51 tokens</li></ul> | <ul><li>0: ~84.20%</li><li>1: ~15.80%</li></ul> |
+* Samples:
+  | doc                                                                                                           | candidate                                                                                                    | label          |
+  |:--------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------|:---------------|
+  | <code>круглое пляжное парео селфи коврик пляжная подстилка пляжное покрывало пляжный коврик пироженко </code> | <code>круглое пляжное парео селфи коврик пляжная подстилка пляжное покрывало пляжный коврик клубника </code> | <code>0</code> |
+  | <code>аккумулятор батарея для ноутбука asus g751 </code>                                                      | <code>аккумулятор батарея для ноутбука asus g75 series</code>                                                | <code>0</code> |
+  | <code>миксер bosch mfq3520 mfq 3520 </code>                                                                   | <code>миксер bosch mfq 4020 </code>                                                                          | <code>0</code> |
+* Loss: [<code>CoSENTLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosentloss) with these parameters:
+  ```json
+  {
+      "scale": 20.0,
+      "similarity_fct": "pairwise_cos_sim"
+  }
+  ```
+
+### Training Hyperparameters
+#### Non-Default Hyperparameters
+
+- `eval_strategy`: steps
+- `per_device_train_batch_size`: 16
+- `per_device_eval_batch_size`: 16
+- `num_train_epochs`: 2
+- `lr_scheduler_type`: cosine
+- `warmup_ratio`: 0.1
+- `load_best_model_at_end`: True
+- `batch_sampler`: no_duplicates
+
+#### All Hyperparameters
+<details><summary>Click to expand</summary>
+
+- `overwrite_output_dir`: False
+- `do_predict`: False
+- `eval_strategy`: steps
+- `prediction_loss_only`: True
+- `per_device_train_batch_size`: 16
+- `per_device_eval_batch_size`: 16
+- `per_gpu_train_batch_size`: None
+- `per_gpu_eval_batch_size`: None
+- `gradient_accumulation_steps`: 1
+- `eval_accumulation_steps`: None
+- `torch_empty_cache_steps`: None
+- `learning_rate`: 5e-05
+- `weight_decay`: 0.0
+- `adam_beta1`: 0.9
+- `adam_beta2`: 0.999
+- `adam_epsilon`: 1e-08
+- `max_grad_norm`: 1.0
+- `num_train_epochs`: 2
+- `max_steps`: -1
+- `lr_scheduler_type`: cosine
+- `lr_scheduler_kwargs`: {}
+- `warmup_ratio`: 0.1
+- `warmup_steps`: 0
+- `log_level`: passive
+- `log_level_replica`: warning
+- `log_on_each_node`: True
+- `logging_nan_inf_filter`: True
+- `save_safetensors`: True
+- `save_on_each_node`: False
+- `save_only_model`: False
+- `restore_callback_states_from_checkpoint`: False
+- `no_cuda`: False
+- `use_cpu`: False
+- `use_mps_device`: False
+- `seed`: 42
+- `data_seed`: None
+- `jit_mode_eval`: False
+- `use_ipex`: False
+- `bf16`: False
+- `fp16`: False
+- `fp16_opt_level`: O1
+- `half_precision_backend`: auto
+- `bf16_full_eval`: False
+- `fp16_full_eval`: False
+- `tf32`: None
+- `local_rank`: 0
+- `ddp_backend`: None
+- `tpu_num_cores`: None
+- `tpu_metrics_debug`: False
+- `debug`: []
+- `dataloader_drop_last`: False
+- `dataloader_num_workers`: 0
+- `dataloader_prefetch_factor`: None
+- `past_index`: -1
+- `disable_tqdm`: False
+- `remove_unused_columns`: True
+- `label_names`: None
+- `load_best_model_at_end`: True
+- `ignore_data_skip`: False
+- `fsdp`: []
+- `fsdp_min_num_params`: 0
+- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
+- `fsdp_transformer_layer_cls_to_wrap`: None
+- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
+- `deepspeed`: None
+- `label_smoothing_factor`: 0.0
+- `optim`: adamw_torch
+- `optim_args`: None
+- `adafactor`: False
+- `group_by_length`: False
+- `length_column_name`: length
+- `ddp_find_unused_parameters`: None
+- `ddp_bucket_cap_mb`: None
+- `ddp_broadcast_buffers`: False
+- `dataloader_pin_memory`: True
+- `dataloader_persistent_workers`: False
+- `skip_memory_metrics`: True
+- `use_legacy_prediction_loop`: False
+- `push_to_hub`: False
+- `resume_from_checkpoint`: None
+- `hub_model_id`: None
+- `hub_strategy`: every_save
+- `hub_private_repo`: False
+- `hub_always_push`: False
+- `gradient_checkpointing`: False
+- `gradient_checkpointing_kwargs`: None
+- `include_inputs_for_metrics`: False
+- `include_for_metrics`: []
+- `eval_do_concat_batches`: True
+- `fp16_backend`: auto
+- `push_to_hub_model_id`: None
+- `push_to_hub_organization`: None
+- `mp_parameters`: 
+- `auto_find_batch_size`: False
+- `full_determinism`: False
+- `torchdynamo`: None
+- `ray_scope`: last
+- `ddp_timeout`: 1800
+- `torch_compile`: False
+- `torch_compile_backend`: None
+- `torch_compile_mode`: None
+- `dispatch_batches`: None
+- `split_batches`: None
+- `include_tokens_per_second`: False
+- `include_num_input_tokens_seen`: False
+- `neftune_noise_alpha`: None
+- `optim_target_modules`: None
+- `batch_eval_metrics`: False
+- `eval_on_start`: False
+- `use_liger_kernel`: False
+- `eval_use_gather_object`: False
+- `average_tokens_across_devices`: False
+- `prompts`: None
+- `batch_sampler`: no_duplicates
+- `multi_dataset_batch_sampler`: proportional
+
+</details>
+
+### Training Logs
+| Epoch  | Step | Training Loss | Validation Loss | example-dev_spearman_cosine |
+|:------:|:----:|:-------------:|:---------------:|:---------------------------:|
+| 0      | 0    | -             | -               | 0.1562                      |
+| 0.1254 | 500  | 4.2363        | 3.5101          | 0.3313                      |
+| 0.2508 | 1000 | 3.0049        | 2.8592          | 0.4536                      |
+| 0.3761 | 1500 | 2.6306        | 2.8977          | 0.4704                      |
+| 0.5015 | 2000 | 2.6472        | 2.6703          | 0.4827                      |
+| 0.6269 | 2500 | 2.6626        | 2.6757          | 0.4837                      |
+| 0.7523 | 3000 | 2.6137        | 2.6397          | 0.4883                      |
+| 0.8776 | 3500 | 2.676         | 2.5394          | 0.4936                      |
+| 1.0030 | 4000 | 2.4997        | 2.5984          | 0.4931                      |
+| 1.1284 | 4500 | 2.4901        | 2.6219          | 0.4946                      |
+| 1.2538 | 5000 | 2.4293        | 2.6319          | 0.4943                      |
+| 1.3791 | 5500 | 2.3914        | 2.7122          | 0.4936                      |
+| 1.5045 | 6000 | 2.465         | 2.6573          | 0.4970                      |
+| 1.6299 | 6500 | 2.5711        | 2.6388          | 0.4965                      |
+| 1.7553 | 7000 | 2.5012        | 2.6323          | 0.4967                      |
+| 1.8806 | 7500 | 2.5775        | 2.6231          | 0.4969                      |
+
+
+### Framework Versions
+- Python: 3.10.14
+- Sentence Transformers: 3.3.1
+- Transformers: 4.46.3
+- PyTorch: 2.4.0
+- Accelerate: 0.34.2
+- Datasets: 3.0.1
+- Tokenizers: 0.20.0
+
+## Citation
+
+### BibTeX
+
+#### Sentence Transformers
+```bibtex
+@inproceedings{reimers-2019-sentence-bert,
+    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
+    author = "Reimers, Nils and Gurevych, Iryna",
+    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
+    month = "11",
+    year = "2019",
+    publisher = "Association for Computational Linguistics",
+    url = "https://arxiv.org/abs/1908.10084",
+}
+```
+
+#### CoSENTLoss
+```bibtex
+@online{kexuefm-8847,
+    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
+    author={Su Jianlin},
+    year={2022},
+    month={Jan},
+    url={https://kexue.fm/archives/8847},
+}
+```
+
+<!--
+## Glossary
+
+*Clearly define terms in order to be accessible across audiences.*
+-->
+
+<!--
+## Model Card Authors
+
+*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
+-->
+
+<!--
+## Model Card Contact
+
+*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
+-->

config.json

@@ -0,0 +1,67 @@
+{
+  "_name_or_path": "jinaai/jina-embeddings-v3",
+  "architectures": [
+    "XLMRobertaLoRA"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "auto_map": {
+    "AutoConfig": "jinaai/xlm-roberta-flash-implementation--configuration_xlm_roberta.XLMRobertaFlashConfig",
+    "AutoModel": "jinaai/xlm-roberta-flash-implementation--modeling_lora.XLMRobertaLoRA",
+    "AutoModelForMaskedLM": "jinaai/xlm-roberta-flash-implementation--modeling_xlm_roberta.XLMRobertaForMaskedLM",
+    "AutoModelForPreTraining": "jinaai/xlm-roberta-flash-implementation--modeling_xlm_roberta.XLMRobertaForPreTraining"
+  },
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "emb_pooler": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "load_trained_adapters": true,
+  "lora_adaptations": [
+    "retrieval.query",
+    "retrieval.passage",
+    "separation",
+    "classification",
+    "text-matching"
+  ],
+  "lora_alpha": 1,
+  "lora_dropout_p": 0.0,
+  "lora_main_params_trainable": false,
+  "lora_rank": 4,
+  "matryoshka_dimensions": [
+    32,
+    64,
+    128,
+    256,
+    512,
+    768,
+    1024
+  ],
+  "max_position_embeddings": 8194,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "rotary",
+  "rotary_emb_base": 20000.0,
+  "task_instructions": {
+    "classification": "",
+    "retrieval.passage": "Represent the document for retrieval: ",
+    "retrieval.query": "Represent the query for retrieving evidence documents: ",
+    "separation": "",
+    "text-matching": ""
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.3",
+  "truncate_dim": null,
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "use_flash_attn": false,
+  "use_reentrant": false,
+  "vocab_size": 250002
+}

config_sentence_transformers.json

@@ -0,0 +1,16 @@
+{
+  "__version__": {
+    "sentence_transformers": "3.3.1",
+    "transformers": "4.46.3",
+    "pytorch": "2.4.0"
+  },
+  "prompts": {
+    "retrieval.query": "Represent the query for retrieving evidence documents: ",
+    "retrieval.passage": "Represent the document for retrieval: ",
+    "separation": "",
+    "classification": "",
+    "text-matching": ""
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

custom_st.py

@@ -0,0 +1,229 @@
+import json
+import logging
+import os
+from io import BytesIO
+from typing import Any, Dict, List, Optional, Tuple, Union
+
+import torch
+from torch import nn
+from transformers import AutoConfig, AutoModel, AutoTokenizer
+
+logger = logging.getLogger(__name__)
+
+
+class Transformer(nn.Module):
+    """Huggingface AutoModel to generate token embeddings.
+    Loads the correct class, e.g. BERT / RoBERTa etc.
+
+    Args:
+        model_name_or_path: Huggingface models name
+            (https://huggingface.co/models)
+        max_seq_length: Truncate any inputs longer than max_seq_length
+        model_args: Keyword arguments passed to the Huggingface
+            Transformers model
+        tokenizer_args: Keyword arguments passed to the Huggingface
+            Transformers tokenizer
+        config_args: Keyword arguments passed to the Huggingface
+            Transformers config
+        cache_dir: Cache dir for Huggingface Transformers to store/load
+            models
+        do_lower_case: If true, lowercases the input (independent if the
+            model is cased or not)
+        tokenizer_name_or_path: Name or path of the tokenizer. When
+            None, then model_name_or_path is used
+    """
+
+    save_in_root: bool = True
+
+    def __init__(
+        self,
+        model_name_or_path: str,
+        max_seq_length: int = None,
+        model_args: Dict[str, Any] = None,
+        tokenizer_args: Dict[str, Any] = None,
+        config_args: Dict[str, Any] = None,
+        cache_dir: str = None,
+        do_lower_case: bool = False,
+        tokenizer_name_or_path: str = None,
+        **kwargs,
+    ) -> None:
+        super().__init__()
+        self.config_keys = ["max_seq_length", "do_lower_case"]
+        self.do_lower_case = do_lower_case
+        if model_args is None:
+            model_args = {}
+        if tokenizer_args is None:
+            tokenizer_args = {}
+        if config_args is None:
+            config_args = {}
+
+        if kwargs.get("backend", "torch") != "torch":
+            logger.warning(
+                f'"jinaai/jina-embeddings-v3" is currently not compatible with the {kwargs["backend"]} backend. '
+                'Continuing with the "torch" backend.'
+            )
+        
+        self.config = AutoConfig.from_pretrained(model_name_or_path, **config_args, cache_dir=cache_dir)
+
+        self._lora_adaptations = self.config.lora_adaptations
+        if (
+            not isinstance(self._lora_adaptations, list)
+            or len(self._lora_adaptations) < 1
+        ):
+            raise ValueError(
+                f"`lora_adaptations` must be a list and contain at least one element"
+            )
+        self._adaptation_map = {
+            name: idx for idx, name in enumerate(self._lora_adaptations)
+        }
+
+        self.default_task = model_args.pop('default_task', None)
+
+        self.auto_model = AutoModel.from_pretrained(model_name_or_path, config=self.config, cache_dir=cache_dir, **model_args)
+
+        if max_seq_length is not None and "model_max_length" not in tokenizer_args:
+            tokenizer_args["model_max_length"] = max_seq_length
+        self.tokenizer = AutoTokenizer.from_pretrained(
+            tokenizer_name_or_path if tokenizer_name_or_path is not None else model_name_or_path,
+            cache_dir=cache_dir,
+            **tokenizer_args,
+        )
+
+        # No max_seq_length set. Try to infer from model
+        if max_seq_length is None:
+            if (
+                hasattr(self.auto_model, "config")
+                and hasattr(self.auto_model.config, "max_position_embeddings")
+                and hasattr(self.tokenizer, "model_max_length")
+            ):
+                max_seq_length = min(self.auto_model.config.max_position_embeddings, self.tokenizer.model_max_length)
+
+        self.max_seq_length = max_seq_length
+
+        if tokenizer_name_or_path is not None:
+            self.auto_model.config.tokenizer_class = self.tokenizer.__class__.__name__
+
+
+    @property
+    def default_task(self):
+        return self._default_task
+
+    @default_task.setter
+    def default_task(self, task: Union[None, str]):
+        self._validate_task(task)
+        self._default_task = task
+        
+
+    def _validate_task(self, task: str):
+        if task and task not in self._lora_adaptations:
+            raise ValueError(
+                f"Unsupported task '{task}'. "
+                f"Supported tasks are: {', '.join(self.config.lora_adaptations)}. "
+                f"Alternatively, don't pass the `task` argument to disable LoRA."
+            )
+
+    def forward(
+        self, features: Dict[str, torch.Tensor], task: Optional[str] = None
+    ) -> Dict[str, torch.Tensor]:
+        """Returns token_embeddings, cls_token"""
+        self._validate_task(task)
+        task = task or self.default_task
+        adapter_mask = None
+        if task:
+            task_id = self._adaptation_map[task]
+            num_examples = features['input_ids'].size(0)
+            adapter_mask = torch.full(
+                (num_examples,), task_id, dtype=torch.int32, device=features['input_ids'].device
+            )
+
+        lora_arguments = (
+            {"adapter_mask": adapter_mask} if adapter_mask is not None else {}
+        )
+        features.pop('prompt_length', None)
+        output_states = self.auto_model.forward(**features, **lora_arguments, return_dict=False)
+        output_tokens = output_states[0]
+        features.update({"token_embeddings": output_tokens, "attention_mask": features["attention_mask"]})
+        return features
+
+    def get_word_embedding_dimension(self) -> int:
+        return self.auto_model.config.hidden_size
+
+    def tokenize(
+        self,
+        texts: Union[List[str], List[dict], List[Tuple[str, str]]],
+        padding: Union[str, bool] = True
+    ) -> Dict[str, torch.Tensor]:
+        """Tokenizes a text and maps tokens to token-ids"""
+        output = {}
+        if isinstance(texts[0], str):
+            to_tokenize = [texts]
+        elif isinstance(texts[0], dict):
+            to_tokenize = []
+            output["text_keys"] = []
+            for lookup in texts:
+                text_key, text = next(iter(lookup.items()))
+                to_tokenize.append(text)
+                output["text_keys"].append(text_key)
+            to_tokenize = [to_tokenize]
+        else:
+            batch1, batch2 = [], []
+            for text_tuple in texts:
+                batch1.append(text_tuple[0])
+                batch2.append(text_tuple[1])
+            to_tokenize = [batch1, batch2]
+
+        # strip
+        to_tokenize = [[str(s).strip() for s in col] for col in to_tokenize]
+
+        # Lowercase
+        if self.do_lower_case:
+            to_tokenize = [[s.lower() for s in col] for col in to_tokenize]
+
+        output.update(
+            self.tokenizer(
+                *to_tokenize,
+                padding=padding,
+                truncation="longest_first",
+                return_tensors="pt",
+                max_length=self.max_seq_length,
+            )
+        )
+        return output
+
+    def get_config_dict(self) -> Dict[str, Any]:
+         return {key: self.__dict__[key] for key in self.config_keys}
+
+    def save(self, output_path: str, safe_serialization: bool = True) -> None:
+        self.auto_model.save_pretrained(output_path, safe_serialization=safe_serialization)
+        self.tokenizer.save_pretrained(output_path)
+
+        with open(os.path.join(output_path, "sentence_bert_config.json"), "w") as fOut:
+            json.dump(self.get_config_dict(), fOut, indent=2)
+
+
+    @classmethod
+    def load(cls, input_path: str) -> "Transformer":
+        # Old classes used other config names than 'sentence_bert_config.json'
+        for config_name in [
+            "sentence_bert_config.json",
+            "sentence_roberta_config.json",
+            "sentence_distilbert_config.json",
+            "sentence_camembert_config.json",
+            "sentence_albert_config.json",
+            "sentence_xlm-roberta_config.json",
+            "sentence_xlnet_config.json",
+        ]:
+            sbert_config_path = os.path.join(input_path, config_name)
+            if os.path.exists(sbert_config_path):
+                break
+
+        with open(sbert_config_path) as fIn:
+            config = json.load(fIn)
+        # Don't allow configs to set trust_remote_code
+        if "model_args" in config and "trust_remote_code" in config["model_args"]:
+            config["model_args"].pop("trust_remote_code")
+        if "tokenizer_args" in config and "trust_remote_code" in config["tokenizer_args"]:
+            config["tokenizer_args"].pop("trust_remote_code")
+        if "config_args" in config and "trust_remote_code" in config["config_args"]:
+            config["config_args"].pop("trust_remote_code")
+        return cls(model_name_or_path=input_path, **config)

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec2a7fec4584bf58064ee4197578edc9993e4e7c6fc6adb03fc57a0088770289
+size 2289306368

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "transformer",
+    "path": "",
+    "type": "custom_st.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "pooler",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "normalizer",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 8194,
+  "do_lower_case": false
+}

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e19cd8c08f528b481e909f73dbd1fd62b1e8b1117579ba205e477801237f9e0
+size 17082988

tokenizer_config.json

@@ -0,0 +1,54 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 8194,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}