File size: 26,185 Bytes
7a50d08 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 |
---
language:
- en
tags:
- ColBERT
- PyLate
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:497901
- loss:Contrastive
base_model: colbert-ir/colbertv2.0
datasets:
- sentence-transformers/msmarco-bm25
pipeline_tag: sentence-similarity
library_name: PyLate
---
# PyLate model based on colbert-ir/colbertv2.0
This is a [PyLate](https://github.com/lightonai/pylate) model finetuned from [colbert-ir/colbertv2.0](https://huggingface.co/colbert-ir/colbertv2.0) on the [msmarco-bm25](https://huggingface.co/datasets/sentence-transformers/msmarco-bm25) dataset. It maps sentences & paragraphs to sequences of 128-dimensional dense vectors and can be used for semantic textual similarity using the MaxSim operator.
## Model Details
### Model Description
- **Model Type:** PyLate model
- **Base model:** [colbert-ir/colbertv2.0](https://huggingface.co/colbert-ir/colbertv2.0) <!-- at revision c1e84128e85ef755c096a95bdb06b47793b13acf -->
- **Document Length:** 180 tokens
- **Query Length:** 32 tokens
- **Output Dimensionality:** 128 tokens
- **Similarity Function:** MaxSim
- **Training Dataset:**
- [msmarco-bm25](https://huggingface.co/datasets/sentence-transformers/msmarco-bm25)
- **Language:** en
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [PyLate Documentation](https://lightonai.github.io/pylate/)
- **Repository:** [PyLate on GitHub](https://github.com/lightonai/pylate)
- **Hugging Face:** [PyLate models on Hugging Face](https://huggingface.co/models?library=PyLate)
### Full Model Architecture
```
ColBERT(
(0): Transformer({'max_seq_length': 179, 'do_lower_case': False}) with Transformer model: BertModel
(1): Dense({'in_features': 768, 'out_features': 128, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
)
```
## Usage
First install the PyLate library:
```bash
pip install -U pylate
```
### Retrieval
PyLate provides a streamlined interface to index and retrieve documents using ColBERT models. The index leverages the Voyager HNSW index to efficiently handle document embeddings and enable fast retrieval.
#### Indexing documents
First, load the ColBERT model and initialize the Voyager index, then encode and index your documents:
```python
from pylate import indexes, models, retrieve
# Step 1: Load the ColBERT model
model = models.ColBERT(
model_name_or_path=pylate_model_id,
)
# Step 2: Initialize the Voyager index
index = indexes.Voyager(
index_folder="pylate-index",
index_name="index",
override=True, # This overwrites the existing index if any
)
# Step 3: Encode the documents
documents_ids = ["1", "2", "3"]
documents = ["document 1 text", "document 2 text", "document 3 text"]
documents_embeddings = model.encode(
documents,
batch_size=32,
is_query=False, # Ensure that it is set to False to indicate that these are documents, not queries
show_progress_bar=True,
)
# Step 4: Add document embeddings to the index by providing embeddings and corresponding ids
index.add_documents(
documents_ids=documents_ids,
documents_embeddings=documents_embeddings,
)
```
Note that you do not have to recreate the index and encode the documents every time. Once you have created an index and added the documents, you can re-use the index later by loading it:
```python
# To load an index, simply instantiate it with the correct folder/name and without overriding it
index = indexes.Voyager(
index_folder="pylate-index",
index_name="index",
)
```
#### Retrieving top-k documents for queries
Once the documents are indexed, you can retrieve the top-k most relevant documents for a given set of queries.
To do so, initialize the ColBERT retriever with the index you want to search in, encode the queries and then retrieve the top-k documents to get the top matches ids and relevance scores:
```python
# Step 1: Initialize the ColBERT retriever
retriever = retrieve.ColBERT(index=index)
# Step 2: Encode the queries
queries_embeddings = model.encode(
["query for document 3", "query for document 1"],
batch_size=32,
is_query=True, # # Ensure that it is set to False to indicate that these are queries
show_progress_bar=True,
)
# Step 3: Retrieve top-k documents
scores = retriever.retrieve(
queries_embeddings=queries_embeddings,
k=10, # Retrieve the top 10 matches for each query
)
```
### Reranking
If you only want to use the ColBERT model to perform reranking on top of your first-stage retrieval pipeline without building an index, you can simply use rank function and pass the queries and documents to rerank:
```python
from pylate import rank, models
queries = [
"query A",
"query B",
]
documents = [
["document A", "document B"],
["document 1", "document C", "document B"],
]
documents_ids = [
[1, 2],
[1, 3, 2],
]
model = models.ColBERT(
model_name_or_path=pylate_model_id,
)
queries_embeddings = model.encode(
queries,
is_query=True,
)
documents_embeddings = model.encode(
documents,
is_query=False,
)
reranked_documents = rank.rerank(
documents_ids=documents_ids,
queries_embeddings=queries_embeddings,
documents_embeddings=documents_embeddings,
)
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### msmarco-bm25
* Dataset: [msmarco-bm25](https://huggingface.co/datasets/sentence-transformers/msmarco-bm25) at [ce8a493](https://huggingface.co/datasets/sentence-transformers/msmarco-bm25/tree/ce8a493a65af5e872c3c92f72a89e2e99e175f02)
* Size: 497,901 training samples
* Columns: <code>query</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
| | query | positive | negative |
|:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | string |
| details | <ul><li>min: 5 tokens</li><li>mean: 10.14 tokens</li><li>max: 20 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 31.91 tokens</li><li>max: 32 tokens</li></ul> | <ul><li>min: 17 tokens</li><li>mean: 31.84 tokens</li><li>max: 32 tokens</li></ul> |
* Samples:
| query | positive | negative |
|:---------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>what is null hypothesis and why is it used in experimental research</code> | <code>A null hypothesis is one that is assumed to be true unless it has been contradicted. It is used to compare to another hypothesis. The experimental hypothesis is what you are observing, and you expect it to differ from the control. erm i know that a null hypothesis is when nothing happens at all i think.</code> | <code>A null hypothesis is one that is assumed to be true unless it has been contradicted. It is used to compare to another hypothesis. The experimental hypothesis is what you are observing, and you expect it to differ from the control. erm i know that a null hypothesis is when nothing happens at all i think.</code> |
| <code>number of students per instructor</code> | <code>The article posited that students preferred classes of 10-20 students, and instructors suggested that the ideal class would have 19 students. Instructors reported that at 39 students problems began to arise, and that a class of 51 students was impossible. They also reported that an uncomfortably small class begins at 7 students, and an impossibly small class has 4 or less.</code> | <code>The ratio of instructors to students isnât as important here as in the lab setting. One to two instructors per 10 students will suffice. Once the students are divided into groups, the instructor should begin to methodically teach ECG interpretation. The instructor should start with waveform definition and recognition.</code> |
| <code>when should exclamation marks be used?</code> | <code>The exclamation mark (British English) or exclamation point (American English) is a punctuation mark usually used after an interjection or exclamation to indicate strong feelings or high volume (shouting), and often marks the end of a sentence.</code> | <code>1 Question marks and exclamation marks go inside the quotation marks when the quoted material is a question or an exclamation and outside the quotation marks when the whole sentence is a question or an exclamation. Question marks and exclamation marks go inside the quotation marks when the quoted material is a question or an exclamation and outside the quotation marks when the whole sentence is a question or an exclamation.</code> |
* Loss: <code>pylate.losses.contrastive.Contrastive</code>
### Evaluation Dataset
#### msmarco-bm25
* Dataset: [msmarco-bm25](https://huggingface.co/datasets/sentence-transformers/msmarco-bm25) at [ce8a493](https://huggingface.co/datasets/sentence-transformers/msmarco-bm25/tree/ce8a493a65af5e872c3c92f72a89e2e99e175f02)
* Size: 5,030 evaluation samples
* Columns: <code>query</code>, <code>positive</code>, and <code>negative</code>
* Approximate statistics based on the first 1000 samples:
| | query | positive | negative |
|:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
| type | string | string | string |
| details | <ul><li>min: 5 tokens</li><li>mean: 10.17 tokens</li><li>max: 32 tokens</li></ul> | <ul><li>min: 20 tokens</li><li>mean: 31.92 tokens</li><li>max: 32 tokens</li></ul> | <ul><li>min: 16 tokens</li><li>mean: 31.93 tokens</li><li>max: 32 tokens</li></ul> |
* Samples:
| query | positive | negative |
|:---------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>what is a hypermarket</code> | <code>By definition a hypermarket is the combination of a supermarket and a department store which has at least 150,000 square feet of floor space, and at least 35% of that space is used for the sale of nonfood merchandise. Generally the terms hypermarket, and superstore are used interchangeably.</code> | <code>hypermarket meaning, definition, what is hypermarket: a very large shop, usually outside the centre of town. Learn more.</code> |
| <code>what is fd&c yellow #6 lake.</code> | <code>FD&C Yellow No. 6 Lake is a color additive used for drug dosage forms such as tablets and capsules. It is also approved for use in foods and cosmetics. FD&C Yellow No. 6 Lake imparts a reddish-yellow color to medicinal dosage forms. FDA performs regulatory review for color additives used in foods, drugs, cosmetics, and medical devices. FD&C specifies the color is approved for use in food, drugs and cosmetics. FD&C Yellow No. 6 Lake may be safely used as a color additive when following FDA specifications. To form lake colors, straight dyes (such as FD&C Yellow No. 6) are mixed with precipitants and salts. Aluminum may be a component. Lakes may be used as color additives for tablet coatings due to their stability.</code> | <code>Coumadin: 6 mg [scored; contains fd&c blue #1 aluminum lake, fd&c yellow #6 aluminum lake] Coumadin: 7.5 mg [scored; contains fd&c yellow #10 aluminum lake, fd&c yellow #6 aluminum lake] Coumadin: 10 mg [scored; dye free] Jantoven: 1 mg [scored; contains fd&c red #40 aluminum lake]</code> |
| <code>how long can ringworm live on clothes</code> | <code>-Sometimes the ringworm on the scalp can causes patches of hair loss. Ringworm in dogs can be spread many of the same ways. Even sharing clothes, towels, or combs may result in spreading the infection. Ringworm is caused by different kinds of fungus on the skin, hair, or nails caused by an infection.he fungus that causes ringworm can typically live up to 7 days on surfaces such as counter tops, carpets, and floors, but it has been reported that some types can live up to one year.</code> | <code>What Causes Ringworm? Ringworm is more common in unsanitary and crowded places. That's because it can live on both skin and surfaces like shower floors, and can be transferred by sharing clothes, sheets, and towels. Even other mammals, including cats and dogs, can easily transfer ringworm to humans. What Are the Types of Ringworm?</code> |
* Loss: <code>pylate.losses.contrastive.Contrastive</code>
### Training Hyperparameters
#### Non-Default Hyperparameters
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `learning_rate`: 3e-06
- `num_train_epochs`: 1
- `fp16`: True
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 32
- `per_device_eval_batch_size`: 32
- `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`: 3e-06
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1.0
- `num_train_epochs`: 1
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: True
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `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`: batch_sampler
- `multi_dataset_batch_sampler`: proportional
</details>
### Training Logs
| Epoch | Step | Training Loss |
|:------:|:-----:|:-------------:|
| 0.0321 | 500 | 0.4976 |
| 0.0643 | 1000 | 0.3532 |
| 0.0964 | 1500 | 0.3195 |
| 0.1285 | 2000 | 0.3079 |
| 0.1607 | 2500 | 0.3067 |
| 0.1928 | 3000 | 0.2957 |
| 0.2249 | 3500 | 0.3086 |
| 0.2571 | 4000 | 0.2927 |
| 0.2892 | 4500 | 0.2922 |
| 0.3213 | 5000 | 0.2931 |
| 0.3535 | 5500 | 0.2957 |
| 0.3856 | 6000 | 0.2809 |
| 0.4177 | 6500 | 0.2773 |
| 0.4499 | 7000 | 0.2728 |
| 0.4820 | 7500 | 0.2888 |
| 0.5141 | 8000 | 0.2863 |
| 0.5463 | 8500 | 0.2813 |
| 0.5784 | 9000 | 0.2695 |
| 0.6105 | 9500 | 0.2834 |
| 0.6427 | 10000 | 0.2739 |
| 0.6748 | 10500 | 0.2744 |
| 0.7069 | 11000 | 0.2849 |
| 0.7391 | 11500 | 0.2808 |
| 0.7712 | 12000 | 0.2796 |
| 0.8033 | 12500 | 0.2772 |
| 0.8355 | 13000 | 0.2813 |
| 0.8676 | 13500 | 0.2756 |
| 0.8997 | 14000 | 0.2771 |
| 0.9319 | 14500 | 0.283 |
| 0.9640 | 15000 | 0.2731 |
| 0.9961 | 15500 | 0.2865 |
### Framework Versions
- Python: 3.12.4
- Sentence Transformers: 4.0.2
- PyLate: 1.2.0
- Transformers: 4.48.2
- PyTorch: 2.6.0+cu124
- Accelerate: 1.7.0
- Datasets: 3.6.0
- Tokenizers: 0.21.1
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084"
}
```
#### PyLate
```bibtex
@misc{PyLate,
title={PyLate: Flexible Training and Retrieval for Late Interaction Models},
author={Chaffin, Antoine and Sourty, Raphaël},
url={https://github.com/lightonai/pylate},
year={2024}
}
```
<!--
## 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.*
--> |