Remove all files to get rid of all files before copying archive over this and adding via LFS
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- checkpoint-1000/1_Pooling/config.json +0 -10
- checkpoint-1000/README.md +0 -466
- checkpoint-1000/config.json +0 -25
- checkpoint-1000/config_sentence_transformers.json +0 -10
- checkpoint-1000/model.safetensors +0 -3
- checkpoint-1000/modules.json +0 -14
- checkpoint-1000/optimizer.pt +0 -3
- checkpoint-1000/rng_state.pth +0 -3
- checkpoint-1000/scaler.pt +0 -3
- checkpoint-1000/scheduler.pt +0 -3
- checkpoint-1000/sentence_bert_config.json +0 -4
- checkpoint-1000/special_tokens_map.json +0 -51
- checkpoint-1000/tokenizer.json +0 -3
- checkpoint-1000/tokenizer_config.json +0 -65
- checkpoint-1000/trainer_state.json +0 -217
- checkpoint-1000/training_args.bin +0 -3
- checkpoint-1000/unigram.json +0 -3
- checkpoint-1060/1_Pooling/config.json +0 -10
- checkpoint-1060/README.md +0 -466
- checkpoint-1060/config.json +0 -25
- checkpoint-1060/config_sentence_transformers.json +0 -10
- checkpoint-1060/model.safetensors +0 -3
- checkpoint-1060/modules.json +0 -14
- checkpoint-1060/optimizer.pt +0 -3
- checkpoint-1060/rng_state.pth +0 -3
- checkpoint-1060/scaler.pt +0 -3
- checkpoint-1060/scheduler.pt +0 -3
- checkpoint-1060/sentence_bert_config.json +0 -4
- checkpoint-1060/special_tokens_map.json +0 -51
- checkpoint-1060/tokenizer.json +0 -3
- checkpoint-1060/tokenizer_config.json +0 -65
- checkpoint-1060/trainer_state.json +0 -217
- checkpoint-1060/training_args.bin +0 -3
- checkpoint-1060/unigram.json +0 -3
- checkpoint-700/1_Pooling/config.json +0 -10
- checkpoint-700/README.md +0 -463
- checkpoint-700/config.json +0 -25
- checkpoint-700/config_sentence_transformers.json +0 -10
- checkpoint-700/model.safetensors +0 -3
- checkpoint-700/modules.json +0 -14
- checkpoint-700/optimizer.pt +0 -3
- checkpoint-700/rng_state.pth +0 -3
- checkpoint-700/scaler.pt +0 -3
- checkpoint-700/scheduler.pt +0 -3
- checkpoint-700/sentence_bert_config.json +0 -4
- checkpoint-700/special_tokens_map.json +0 -51
- checkpoint-700/tokenizer.json +0 -3
- checkpoint-700/tokenizer_config.json +0 -65
- checkpoint-700/trainer_state.json +0 -162
- checkpoint-700/training_args.bin +0 -3
checkpoint-1000/1_Pooling/config.json
DELETED
|
@@ -1,10 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"word_embedding_dimension": 384,
|
| 3 |
-
"pooling_mode_cls_token": false,
|
| 4 |
-
"pooling_mode_mean_tokens": true,
|
| 5 |
-
"pooling_mode_max_tokens": false,
|
| 6 |
-
"pooling_mode_mean_sqrt_len_tokens": false,
|
| 7 |
-
"pooling_mode_weightedmean_tokens": false,
|
| 8 |
-
"pooling_mode_lasttoken": false,
|
| 9 |
-
"include_prompt": true
|
| 10 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/README.md
DELETED
|
@@ -1,466 +0,0 @@
|
|
| 1 |
-
---
|
| 2 |
-
language:
|
| 3 |
-
- en
|
| 4 |
-
license: apache-2.0
|
| 5 |
-
tags:
|
| 6 |
-
- sentence-transformers
|
| 7 |
-
- sentence-similarity
|
| 8 |
-
- feature-extraction
|
| 9 |
-
- generated_from_trainer
|
| 10 |
-
- dataset_size:2130620
|
| 11 |
-
- loss:ContrastiveLoss
|
| 12 |
-
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
| 13 |
-
widget:
|
| 14 |
-
- source_sentence: مانوئلا دی سنتا
|
| 15 |
-
sentences:
|
| 16 |
-
- Renko Kitagawa
|
| 17 |
-
- هانس هيرمان وير
|
| 18 |
-
- Ди Чента, Мануэла
|
| 19 |
-
- source_sentence: يورى جافريلوف
|
| 20 |
-
sentences:
|
| 21 |
-
- Wiktor Pinczuk
|
| 22 |
-
- Natalia Germanovna DIRKS
|
| 23 |
-
- Світлана Євгенівна Савицька
|
| 24 |
-
- source_sentence: Џуди Колинс
|
| 25 |
-
sentences:
|
| 26 |
-
- Collins
|
| 27 |
-
- Aisha Muhammed Abdul Salam
|
| 28 |
-
- Phonic Boy On Dope
|
| 29 |
-
- source_sentence: ויליאם בלייר
|
| 30 |
-
sentences:
|
| 31 |
-
- The Hon. Mr Justice Blair
|
| 32 |
-
- Queen Ingrid of Denmark
|
| 33 |
-
- Herman van Rompuy
|
| 34 |
-
- source_sentence: Saif al-Arab GADAFI
|
| 35 |
-
sentences:
|
| 36 |
-
- Максім Недасекаў
|
| 37 |
-
- Mervyn Allister King
|
| 38 |
-
- Paul d. scully-power
|
| 39 |
-
pipeline_tag: sentence-similarity
|
| 40 |
-
library_name: sentence-transformers
|
| 41 |
-
metrics:
|
| 42 |
-
- cosine_accuracy
|
| 43 |
-
- cosine_accuracy_threshold
|
| 44 |
-
- cosine_f1
|
| 45 |
-
- cosine_f1_threshold
|
| 46 |
-
- cosine_precision
|
| 47 |
-
- cosine_recall
|
| 48 |
-
- cosine_ap
|
| 49 |
-
- cosine_mcc
|
| 50 |
-
model-index:
|
| 51 |
-
- name: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2-address-matcher-original
|
| 52 |
-
results:
|
| 53 |
-
- task:
|
| 54 |
-
type: binary-classification
|
| 55 |
-
name: Binary Classification
|
| 56 |
-
dataset:
|
| 57 |
-
name: sentence transformers paraphrase multilingual MiniLM L12 v2
|
| 58 |
-
type: sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2
|
| 59 |
-
metrics:
|
| 60 |
-
- type: cosine_accuracy
|
| 61 |
-
value: 0.9905380542935456
|
| 62 |
-
name: Cosine Accuracy
|
| 63 |
-
- type: cosine_accuracy_threshold
|
| 64 |
-
value: 0.6790644526481628
|
| 65 |
-
name: Cosine Accuracy Threshold
|
| 66 |
-
- type: cosine_f1
|
| 67 |
-
value: 0.9856131536880567
|
| 68 |
-
name: Cosine F1
|
| 69 |
-
- type: cosine_f1_threshold
|
| 70 |
-
value: 0.6790644526481628
|
| 71 |
-
name: Cosine F1 Threshold
|
| 72 |
-
- type: cosine_precision
|
| 73 |
-
value: 0.9816899806664392
|
| 74 |
-
name: Cosine Precision
|
| 75 |
-
- type: cosine_recall
|
| 76 |
-
value: 0.9895678092399404
|
| 77 |
-
name: Cosine Recall
|
| 78 |
-
- type: cosine_ap
|
| 79 |
-
value: 0.9977983578816215
|
| 80 |
-
name: Cosine Ap
|
| 81 |
-
- type: cosine_mcc
|
| 82 |
-
value: 0.9785817179348335
|
| 83 |
-
name: Cosine Mcc
|
| 84 |
-
---
|
| 85 |
-
|
| 86 |
-
# sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2-address-matcher-original
|
| 87 |
-
|
| 88 |
-
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
| 89 |
-
|
| 90 |
-
## Model Details
|
| 91 |
-
|
| 92 |
-
### Model Description
|
| 93 |
-
- **Model Type:** Sentence Transformer
|
| 94 |
-
- **Base model:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) <!-- at revision 86741b4e3f5cb7765a600d3a3d55a0f6a6cb443d -->
|
| 95 |
-
- **Maximum Sequence Length:** 128 tokens
|
| 96 |
-
- **Output Dimensionality:** 384 dimensions
|
| 97 |
-
- **Similarity Function:** Cosine Similarity
|
| 98 |
-
<!-- - **Training Dataset:** Unknown -->
|
| 99 |
-
- **Language:** en
|
| 100 |
-
- **License:** apache-2.0
|
| 101 |
-
|
| 102 |
-
### Model Sources
|
| 103 |
-
|
| 104 |
-
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
| 105 |
-
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
|
| 106 |
-
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
| 107 |
-
|
| 108 |
-
### Full Model Architecture
|
| 109 |
-
|
| 110 |
-
```
|
| 111 |
-
SentenceTransformer(
|
| 112 |
-
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel
|
| 113 |
-
(1): Pooling({'word_embedding_dimension': 384, '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})
|
| 114 |
-
)
|
| 115 |
-
```
|
| 116 |
-
|
| 117 |
-
## Usage
|
| 118 |
-
|
| 119 |
-
### Direct Usage (Sentence Transformers)
|
| 120 |
-
|
| 121 |
-
First install the Sentence Transformers library:
|
| 122 |
-
|
| 123 |
-
```bash
|
| 124 |
-
pip install -U sentence-transformers
|
| 125 |
-
```
|
| 126 |
-
|
| 127 |
-
Then you can load this model and run inference.
|
| 128 |
-
```python
|
| 129 |
-
from sentence_transformers import SentenceTransformer
|
| 130 |
-
|
| 131 |
-
# Download from the 🤗 Hub
|
| 132 |
-
model = SentenceTransformer("sentence_transformers_model_id")
|
| 133 |
-
# Run inference
|
| 134 |
-
sentences = [
|
| 135 |
-
'Saif al-Arab GADAFI',
|
| 136 |
-
'Максім Недасекаў',
|
| 137 |
-
'Mervyn Allister King',
|
| 138 |
-
]
|
| 139 |
-
embeddings = model.encode(sentences)
|
| 140 |
-
print(embeddings.shape)
|
| 141 |
-
# [3, 384]
|
| 142 |
-
|
| 143 |
-
# Get the similarity scores for the embeddings
|
| 144 |
-
similarities = model.similarity(embeddings, embeddings)
|
| 145 |
-
print(similarities.shape)
|
| 146 |
-
# [3, 3]
|
| 147 |
-
```
|
| 148 |
-
|
| 149 |
-
<!--
|
| 150 |
-
### Direct Usage (Transformers)
|
| 151 |
-
|
| 152 |
-
<details><summary>Click to see the direct usage in Transformers</summary>
|
| 153 |
-
|
| 154 |
-
</details>
|
| 155 |
-
-->
|
| 156 |
-
|
| 157 |
-
<!--
|
| 158 |
-
### Downstream Usage (Sentence Transformers)
|
| 159 |
-
|
| 160 |
-
You can finetune this model on your own dataset.
|
| 161 |
-
|
| 162 |
-
<details><summary>Click to expand</summary>
|
| 163 |
-
|
| 164 |
-
</details>
|
| 165 |
-
-->
|
| 166 |
-
|
| 167 |
-
<!--
|
| 168 |
-
### Out-of-Scope Use
|
| 169 |
-
|
| 170 |
-
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
| 171 |
-
-->
|
| 172 |
-
|
| 173 |
-
## Evaluation
|
| 174 |
-
|
| 175 |
-
### Metrics
|
| 176 |
-
|
| 177 |
-
#### Binary Classification
|
| 178 |
-
|
| 179 |
-
* Dataset: `sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2`
|
| 180 |
-
* Evaluated with [<code>BinaryClassificationEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator)
|
| 181 |
-
|
| 182 |
-
| Metric | Value |
|
| 183 |
-
|:--------------------------|:-----------|
|
| 184 |
-
| cosine_accuracy | 0.9905 |
|
| 185 |
-
| cosine_accuracy_threshold | 0.6791 |
|
| 186 |
-
| cosine_f1 | 0.9856 |
|
| 187 |
-
| cosine_f1_threshold | 0.6791 |
|
| 188 |
-
| cosine_precision | 0.9817 |
|
| 189 |
-
| cosine_recall | 0.9896 |
|
| 190 |
-
| **cosine_ap** | **0.9978** |
|
| 191 |
-
| cosine_mcc | 0.9786 |
|
| 192 |
-
|
| 193 |
-
<!--
|
| 194 |
-
## Bias, Risks and Limitations
|
| 195 |
-
|
| 196 |
-
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
| 197 |
-
-->
|
| 198 |
-
|
| 199 |
-
<!--
|
| 200 |
-
### Recommendations
|
| 201 |
-
|
| 202 |
-
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
| 203 |
-
-->
|
| 204 |
-
|
| 205 |
-
## Training Details
|
| 206 |
-
|
| 207 |
-
### Training Dataset
|
| 208 |
-
|
| 209 |
-
#### Unnamed Dataset
|
| 210 |
-
|
| 211 |
-
* Size: 2,130,620 training samples
|
| 212 |
-
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 213 |
-
* Approximate statistics based on the first 1000 samples:
|
| 214 |
-
| | sentence1 | sentence2 | label |
|
| 215 |
-
|:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 216 |
-
| type | string | string | float |
|
| 217 |
-
| details | <ul><li>min: 3 tokens</li><li>mean: 9.28 tokens</li><li>max: 57 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 9.11 tokens</li><li>max: 65 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.34</li><li>max: 1.0</li></ul> |
|
| 218 |
-
* Samples:
|
| 219 |
-
| sentence1 | sentence2 | label |
|
| 220 |
-
|:----------------------------|:-------------------------------|:-----------------|
|
| 221 |
-
| <code>ג'ק וייט</code> | <code>Jack White</code> | <code>1.0</code> |
|
| 222 |
-
| <code>Абдуллоҳ Гул</code> | <code>Савицкая Светлана</code> | <code>0.0</code> |
|
| 223 |
-
| <code>ショーン・ジャスティン・ペン</code> | <code>شان پن</code> | <code>1.0</code> |
|
| 224 |
-
* Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters:
|
| 225 |
-
```json
|
| 226 |
-
{
|
| 227 |
-
"distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
|
| 228 |
-
"margin": 0.5,
|
| 229 |
-
"size_average": true
|
| 230 |
-
}
|
| 231 |
-
```
|
| 232 |
-
|
| 233 |
-
### Evaluation Dataset
|
| 234 |
-
|
| 235 |
-
#### Unnamed Dataset
|
| 236 |
-
|
| 237 |
-
* Size: 266,328 evaluation samples
|
| 238 |
-
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 239 |
-
* Approximate statistics based on the first 1000 samples:
|
| 240 |
-
| | sentence1 | sentence2 | label |
|
| 241 |
-
|:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 242 |
-
| type | string | string | float |
|
| 243 |
-
| details | <ul><li>min: 3 tokens</li><li>mean: 9.27 tokens</li><li>max: 79 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 8.99 tokens</li><li>max: 61 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.32</li><li>max: 1.0</li></ul> |
|
| 244 |
-
* Samples:
|
| 245 |
-
| sentence1 | sentence2 | label |
|
| 246 |
-
|:---------------------------------------------|:-----------------------------------------------|:-----------------|
|
| 247 |
-
| <code>Анатолий Николаевич Герасимов</code> | <code>Anatoli Nikolajewitsch Gerassimow</code> | <code>1.0</code> |
|
| 248 |
-
| <code>Igor Stanislavovitsj Prokopenko</code> | <code>Angelo Lauricella</code> | <code>0.0</code> |
|
| 249 |
-
| <code>Кофе, Линда</code> | <code>Святлана Яўгенаўна Савіцкая</code> | <code>0.0</code> |
|
| 250 |
-
* Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters:
|
| 251 |
-
```json
|
| 252 |
-
{
|
| 253 |
-
"distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
|
| 254 |
-
"margin": 0.5,
|
| 255 |
-
"size_average": true
|
| 256 |
-
}
|
| 257 |
-
```
|
| 258 |
-
|
| 259 |
-
### Training Hyperparameters
|
| 260 |
-
#### Non-Default Hyperparameters
|
| 261 |
-
|
| 262 |
-
- `eval_strategy`: steps
|
| 263 |
-
- `per_device_train_batch_size`: 5000
|
| 264 |
-
- `per_device_eval_batch_size`: 5000
|
| 265 |
-
- `gradient_accumulation_steps`: 4
|
| 266 |
-
- `weight_decay`: 0.02
|
| 267 |
-
- `num_train_epochs`: 10
|
| 268 |
-
- `warmup_ratio`: 0.1
|
| 269 |
-
- `fp16`: True
|
| 270 |
-
- `load_best_model_at_end`: True
|
| 271 |
-
- `optim`: adafactor
|
| 272 |
-
- `gradient_checkpointing`: True
|
| 273 |
-
|
| 274 |
-
#### All Hyperparameters
|
| 275 |
-
<details><summary>Click to expand</summary>
|
| 276 |
-
|
| 277 |
-
- `overwrite_output_dir`: False
|
| 278 |
-
- `do_predict`: False
|
| 279 |
-
- `eval_strategy`: steps
|
| 280 |
-
- `prediction_loss_only`: True
|
| 281 |
-
- `per_device_train_batch_size`: 5000
|
| 282 |
-
- `per_device_eval_batch_size`: 5000
|
| 283 |
-
- `per_gpu_train_batch_size`: None
|
| 284 |
-
- `per_gpu_eval_batch_size`: None
|
| 285 |
-
- `gradient_accumulation_steps`: 4
|
| 286 |
-
- `eval_accumulation_steps`: None
|
| 287 |
-
- `torch_empty_cache_steps`: None
|
| 288 |
-
- `learning_rate`: 5e-05
|
| 289 |
-
- `weight_decay`: 0.02
|
| 290 |
-
- `adam_beta1`: 0.9
|
| 291 |
-
- `adam_beta2`: 0.999
|
| 292 |
-
- `adam_epsilon`: 1e-08
|
| 293 |
-
- `max_grad_norm`: 1.0
|
| 294 |
-
- `num_train_epochs`: 10
|
| 295 |
-
- `max_steps`: -1
|
| 296 |
-
- `lr_scheduler_type`: linear
|
| 297 |
-
- `lr_scheduler_kwargs`: {}
|
| 298 |
-
- `warmup_ratio`: 0.1
|
| 299 |
-
- `warmup_steps`: 0
|
| 300 |
-
- `log_level`: passive
|
| 301 |
-
- `log_level_replica`: warning
|
| 302 |
-
- `log_on_each_node`: True
|
| 303 |
-
- `logging_nan_inf_filter`: True
|
| 304 |
-
- `save_safetensors`: True
|
| 305 |
-
- `save_on_each_node`: False
|
| 306 |
-
- `save_only_model`: False
|
| 307 |
-
- `restore_callback_states_from_checkpoint`: False
|
| 308 |
-
- `no_cuda`: False
|
| 309 |
-
- `use_cpu`: False
|
| 310 |
-
- `use_mps_device`: False
|
| 311 |
-
- `seed`: 42
|
| 312 |
-
- `data_seed`: None
|
| 313 |
-
- `jit_mode_eval`: False
|
| 314 |
-
- `use_ipex`: False
|
| 315 |
-
- `bf16`: False
|
| 316 |
-
- `fp16`: True
|
| 317 |
-
- `fp16_opt_level`: O1
|
| 318 |
-
- `half_precision_backend`: auto
|
| 319 |
-
- `bf16_full_eval`: False
|
| 320 |
-
- `fp16_full_eval`: False
|
| 321 |
-
- `tf32`: None
|
| 322 |
-
- `local_rank`: 0
|
| 323 |
-
- `ddp_backend`: None
|
| 324 |
-
- `tpu_num_cores`: None
|
| 325 |
-
- `tpu_metrics_debug`: False
|
| 326 |
-
- `debug`: []
|
| 327 |
-
- `dataloader_drop_last`: False
|
| 328 |
-
- `dataloader_num_workers`: 0
|
| 329 |
-
- `dataloader_prefetch_factor`: None
|
| 330 |
-
- `past_index`: -1
|
| 331 |
-
- `disable_tqdm`: False
|
| 332 |
-
- `remove_unused_columns`: True
|
| 333 |
-
- `label_names`: None
|
| 334 |
-
- `load_best_model_at_end`: True
|
| 335 |
-
- `ignore_data_skip`: False
|
| 336 |
-
- `fsdp`: []
|
| 337 |
-
- `fsdp_min_num_params`: 0
|
| 338 |
-
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
|
| 339 |
-
- `tp_size`: 0
|
| 340 |
-
- `fsdp_transformer_layer_cls_to_wrap`: None
|
| 341 |
-
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
| 342 |
-
- `deepspeed`: None
|
| 343 |
-
- `label_smoothing_factor`: 0.0
|
| 344 |
-
- `optim`: adafactor
|
| 345 |
-
- `optim_args`: None
|
| 346 |
-
- `adafactor`: False
|
| 347 |
-
- `group_by_length`: False
|
| 348 |
-
- `length_column_name`: length
|
| 349 |
-
- `ddp_find_unused_parameters`: None
|
| 350 |
-
- `ddp_bucket_cap_mb`: None
|
| 351 |
-
- `ddp_broadcast_buffers`: False
|
| 352 |
-
- `dataloader_pin_memory`: True
|
| 353 |
-
- `dataloader_persistent_workers`: False
|
| 354 |
-
- `skip_memory_metrics`: True
|
| 355 |
-
- `use_legacy_prediction_loop`: False
|
| 356 |
-
- `push_to_hub`: False
|
| 357 |
-
- `resume_from_checkpoint`: None
|
| 358 |
-
- `hub_model_id`: None
|
| 359 |
-
- `hub_strategy`: every_save
|
| 360 |
-
- `hub_private_repo`: None
|
| 361 |
-
- `hub_always_push`: False
|
| 362 |
-
- `gradient_checkpointing`: True
|
| 363 |
-
- `gradient_checkpointing_kwargs`: None
|
| 364 |
-
- `include_inputs_for_metrics`: False
|
| 365 |
-
- `include_for_metrics`: []
|
| 366 |
-
- `eval_do_concat_batches`: True
|
| 367 |
-
- `fp16_backend`: auto
|
| 368 |
-
- `push_to_hub_model_id`: None
|
| 369 |
-
- `push_to_hub_organization`: None
|
| 370 |
-
- `mp_parameters`:
|
| 371 |
-
- `auto_find_batch_size`: False
|
| 372 |
-
- `full_determinism`: False
|
| 373 |
-
- `torchdynamo`: None
|
| 374 |
-
- `ray_scope`: last
|
| 375 |
-
- `ddp_timeout`: 1800
|
| 376 |
-
- `torch_compile`: False
|
| 377 |
-
- `torch_compile_backend`: None
|
| 378 |
-
- `torch_compile_mode`: None
|
| 379 |
-
- `include_tokens_per_second`: False
|
| 380 |
-
- `include_num_input_tokens_seen`: False
|
| 381 |
-
- `neftune_noise_alpha`: None
|
| 382 |
-
- `optim_target_modules`: None
|
| 383 |
-
- `batch_eval_metrics`: False
|
| 384 |
-
- `eval_on_start`: False
|
| 385 |
-
- `use_liger_kernel`: False
|
| 386 |
-
- `eval_use_gather_object`: False
|
| 387 |
-
- `average_tokens_across_devices`: False
|
| 388 |
-
- `prompts`: None
|
| 389 |
-
- `batch_sampler`: batch_sampler
|
| 390 |
-
- `multi_dataset_batch_sampler`: proportional
|
| 391 |
-
|
| 392 |
-
</details>
|
| 393 |
-
|
| 394 |
-
### Training Logs
|
| 395 |
-
| Epoch | Step | Training Loss | Validation Loss | sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap |
|
| 396 |
-
|:------:|:----:|:-------------:|:---------------:|:---------------------------------------------------------------------:|
|
| 397 |
-
| -1 | -1 | - | - | 0.7195 |
|
| 398 |
-
| 0.9368 | 100 | - | 0.0083 | 0.9597 |
|
| 399 |
-
| 1.8712 | 200 | - | 0.0043 | 0.9877 |
|
| 400 |
-
| 2.8056 | 300 | - | 0.0028 | 0.9936 |
|
| 401 |
-
| 3.7400 | 400 | - | 0.0021 | 0.9954 |
|
| 402 |
-
| 4.6745 | 500 | 0.0224 | 0.0016 | 0.9964 |
|
| 403 |
-
| 5.6089 | 600 | - | 0.0015 | 0.9970 |
|
| 404 |
-
| 6.5433 | 700 | - | 0.0014 | 0.9974 |
|
| 405 |
-
| 7.4778 | 800 | - | 0.0013 | 0.9975 |
|
| 406 |
-
| 8.4122 | 900 | - | 0.0013 | 0.9977 |
|
| 407 |
-
| 9.3466 | 1000 | 0.0052 | 0.0012 | 0.9978 |
|
| 408 |
-
|
| 409 |
-
|
| 410 |
-
### Framework Versions
|
| 411 |
-
- Python: 3.12.9
|
| 412 |
-
- Sentence Transformers: 3.4.1
|
| 413 |
-
- Transformers: 4.51.3
|
| 414 |
-
- PyTorch: 2.7.0+cu126
|
| 415 |
-
- Accelerate: 1.6.0
|
| 416 |
-
- Datasets: 3.6.0
|
| 417 |
-
- Tokenizers: 0.21.1
|
| 418 |
-
|
| 419 |
-
## Citation
|
| 420 |
-
|
| 421 |
-
### BibTeX
|
| 422 |
-
|
| 423 |
-
#### Sentence Transformers
|
| 424 |
-
```bibtex
|
| 425 |
-
@inproceedings{reimers-2019-sentence-bert,
|
| 426 |
-
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
| 427 |
-
author = "Reimers, Nils and Gurevych, Iryna",
|
| 428 |
-
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
| 429 |
-
month = "11",
|
| 430 |
-
year = "2019",
|
| 431 |
-
publisher = "Association for Computational Linguistics",
|
| 432 |
-
url = "https://arxiv.org/abs/1908.10084",
|
| 433 |
-
}
|
| 434 |
-
```
|
| 435 |
-
|
| 436 |
-
#### ContrastiveLoss
|
| 437 |
-
```bibtex
|
| 438 |
-
@inproceedings{hadsell2006dimensionality,
|
| 439 |
-
author={Hadsell, R. and Chopra, S. and LeCun, Y.},
|
| 440 |
-
booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
|
| 441 |
-
title={Dimensionality Reduction by Learning an Invariant Mapping},
|
| 442 |
-
year={2006},
|
| 443 |
-
volume={2},
|
| 444 |
-
number={},
|
| 445 |
-
pages={1735-1742},
|
| 446 |
-
doi={10.1109/CVPR.2006.100}
|
| 447 |
-
}
|
| 448 |
-
```
|
| 449 |
-
|
| 450 |
-
<!--
|
| 451 |
-
## Glossary
|
| 452 |
-
|
| 453 |
-
*Clearly define terms in order to be accessible across audiences.*
|
| 454 |
-
-->
|
| 455 |
-
|
| 456 |
-
<!--
|
| 457 |
-
## Model Card Authors
|
| 458 |
-
|
| 459 |
-
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
| 460 |
-
-->
|
| 461 |
-
|
| 462 |
-
<!--
|
| 463 |
-
## Model Card Contact
|
| 464 |
-
|
| 465 |
-
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
| 466 |
-
-->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/config.json
DELETED
|
@@ -1,25 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"BertModel"
|
| 4 |
-
],
|
| 5 |
-
"attention_probs_dropout_prob": 0.1,
|
| 6 |
-
"classifier_dropout": null,
|
| 7 |
-
"gradient_checkpointing": false,
|
| 8 |
-
"hidden_act": "gelu",
|
| 9 |
-
"hidden_dropout_prob": 0.1,
|
| 10 |
-
"hidden_size": 384,
|
| 11 |
-
"initializer_range": 0.02,
|
| 12 |
-
"intermediate_size": 1536,
|
| 13 |
-
"layer_norm_eps": 1e-12,
|
| 14 |
-
"max_position_embeddings": 512,
|
| 15 |
-
"model_type": "bert",
|
| 16 |
-
"num_attention_heads": 12,
|
| 17 |
-
"num_hidden_layers": 12,
|
| 18 |
-
"pad_token_id": 0,
|
| 19 |
-
"position_embedding_type": "absolute",
|
| 20 |
-
"torch_dtype": "float32",
|
| 21 |
-
"transformers_version": "4.51.3",
|
| 22 |
-
"type_vocab_size": 2,
|
| 23 |
-
"use_cache": true,
|
| 24 |
-
"vocab_size": 250037
|
| 25 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/config_sentence_transformers.json
DELETED
|
@@ -1,10 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"__version__": {
|
| 3 |
-
"sentence_transformers": "3.4.1",
|
| 4 |
-
"transformers": "4.51.3",
|
| 5 |
-
"pytorch": "2.7.0+cu126"
|
| 6 |
-
},
|
| 7 |
-
"prompts": {},
|
| 8 |
-
"default_prompt_name": null,
|
| 9 |
-
"similarity_fn_name": "cosine"
|
| 10 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:4f4d32a2eafc839cb2ab10b136bf98c4d30bdad7f85e5f55ceafdf3a54a9e859
|
| 3 |
-
size 470637416
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/modules.json
DELETED
|
@@ -1,14 +0,0 @@
|
|
| 1 |
-
[
|
| 2 |
-
{
|
| 3 |
-
"idx": 0,
|
| 4 |
-
"name": "0",
|
| 5 |
-
"path": "",
|
| 6 |
-
"type": "sentence_transformers.models.Transformer"
|
| 7 |
-
},
|
| 8 |
-
{
|
| 9 |
-
"idx": 1,
|
| 10 |
-
"name": "1",
|
| 11 |
-
"path": "1_Pooling",
|
| 12 |
-
"type": "sentence_transformers.models.Pooling"
|
| 13 |
-
}
|
| 14 |
-
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/optimizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:c78ab330430a2994c0f6654ccbded4a6558ca0f6cfe08b4fa75960dd3563c6fa
|
| 3 |
-
size 1715019
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:f3ba9bdd5b024d60bbe4f0e967c35c5c47d5c8b8d992558db327a2aae780abce
|
| 3 |
-
size 14645
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/scaler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:14ae2a2128444abab378aa06c09a61a84665f758fcc19fc46f5789b0bc1b5665
|
| 3 |
-
size 1383
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:c6cb8ef30177a351efb4472a87f05bae30f484d65c37957d4e17a50e58c3b3e9
|
| 3 |
-
size 1465
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/sentence_bert_config.json
DELETED
|
@@ -1,4 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"max_seq_length": 128,
|
| 3 |
-
"do_lower_case": false
|
| 4 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/special_tokens_map.json
DELETED
|
@@ -1,51 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"bos_token": {
|
| 3 |
-
"content": "<s>",
|
| 4 |
-
"lstrip": false,
|
| 5 |
-
"normalized": false,
|
| 6 |
-
"rstrip": false,
|
| 7 |
-
"single_word": false
|
| 8 |
-
},
|
| 9 |
-
"cls_token": {
|
| 10 |
-
"content": "<s>",
|
| 11 |
-
"lstrip": false,
|
| 12 |
-
"normalized": false,
|
| 13 |
-
"rstrip": false,
|
| 14 |
-
"single_word": false
|
| 15 |
-
},
|
| 16 |
-
"eos_token": {
|
| 17 |
-
"content": "</s>",
|
| 18 |
-
"lstrip": false,
|
| 19 |
-
"normalized": false,
|
| 20 |
-
"rstrip": false,
|
| 21 |
-
"single_word": false
|
| 22 |
-
},
|
| 23 |
-
"mask_token": {
|
| 24 |
-
"content": "<mask>",
|
| 25 |
-
"lstrip": true,
|
| 26 |
-
"normalized": false,
|
| 27 |
-
"rstrip": false,
|
| 28 |
-
"single_word": false
|
| 29 |
-
},
|
| 30 |
-
"pad_token": {
|
| 31 |
-
"content": "<pad>",
|
| 32 |
-
"lstrip": false,
|
| 33 |
-
"normalized": false,
|
| 34 |
-
"rstrip": false,
|
| 35 |
-
"single_word": false
|
| 36 |
-
},
|
| 37 |
-
"sep_token": {
|
| 38 |
-
"content": "</s>",
|
| 39 |
-
"lstrip": false,
|
| 40 |
-
"normalized": false,
|
| 41 |
-
"rstrip": false,
|
| 42 |
-
"single_word": false
|
| 43 |
-
},
|
| 44 |
-
"unk_token": {
|
| 45 |
-
"content": "<unk>",
|
| 46 |
-
"lstrip": false,
|
| 47 |
-
"normalized": false,
|
| 48 |
-
"rstrip": false,
|
| 49 |
-
"single_word": false
|
| 50 |
-
}
|
| 51 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/tokenizer.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:cad551d5600a84242d0973327029452a1e3672ba6313c2a3c3d69c4310e12719
|
| 3 |
-
size 17082987
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/tokenizer_config.json
DELETED
|
@@ -1,65 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"added_tokens_decoder": {
|
| 3 |
-
"0": {
|
| 4 |
-
"content": "<s>",
|
| 5 |
-
"lstrip": false,
|
| 6 |
-
"normalized": false,
|
| 7 |
-
"rstrip": false,
|
| 8 |
-
"single_word": false,
|
| 9 |
-
"special": true
|
| 10 |
-
},
|
| 11 |
-
"1": {
|
| 12 |
-
"content": "<pad>",
|
| 13 |
-
"lstrip": false,
|
| 14 |
-
"normalized": false,
|
| 15 |
-
"rstrip": false,
|
| 16 |
-
"single_word": false,
|
| 17 |
-
"special": true
|
| 18 |
-
},
|
| 19 |
-
"2": {
|
| 20 |
-
"content": "</s>",
|
| 21 |
-
"lstrip": false,
|
| 22 |
-
"normalized": false,
|
| 23 |
-
"rstrip": false,
|
| 24 |
-
"single_word": false,
|
| 25 |
-
"special": true
|
| 26 |
-
},
|
| 27 |
-
"3": {
|
| 28 |
-
"content": "<unk>",
|
| 29 |
-
"lstrip": false,
|
| 30 |
-
"normalized": false,
|
| 31 |
-
"rstrip": false,
|
| 32 |
-
"single_word": false,
|
| 33 |
-
"special": true
|
| 34 |
-
},
|
| 35 |
-
"250001": {
|
| 36 |
-
"content": "<mask>",
|
| 37 |
-
"lstrip": true,
|
| 38 |
-
"normalized": false,
|
| 39 |
-
"rstrip": false,
|
| 40 |
-
"single_word": false,
|
| 41 |
-
"special": true
|
| 42 |
-
}
|
| 43 |
-
},
|
| 44 |
-
"bos_token": "<s>",
|
| 45 |
-
"clean_up_tokenization_spaces": false,
|
| 46 |
-
"cls_token": "<s>",
|
| 47 |
-
"do_lower_case": true,
|
| 48 |
-
"eos_token": "</s>",
|
| 49 |
-
"extra_special_tokens": {},
|
| 50 |
-
"mask_token": "<mask>",
|
| 51 |
-
"max_length": 128,
|
| 52 |
-
"model_max_length": 128,
|
| 53 |
-
"pad_to_multiple_of": null,
|
| 54 |
-
"pad_token": "<pad>",
|
| 55 |
-
"pad_token_type_id": 0,
|
| 56 |
-
"padding_side": "right",
|
| 57 |
-
"sep_token": "</s>",
|
| 58 |
-
"stride": 0,
|
| 59 |
-
"strip_accents": null,
|
| 60 |
-
"tokenize_chinese_chars": true,
|
| 61 |
-
"tokenizer_class": "BertTokenizer",
|
| 62 |
-
"truncation_side": "right",
|
| 63 |
-
"truncation_strategy": "longest_first",
|
| 64 |
-
"unk_token": "<unk>"
|
| 65 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/trainer_state.json
DELETED
|
@@ -1,217 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": 1000,
|
| 3 |
-
"best_metric": 0.0012360884575173259,
|
| 4 |
-
"best_model_checkpoint": "data/fine-tuned-sbert-sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2-original-adafactor/checkpoint-1000",
|
| 5 |
-
"epoch": 9.346604215456674,
|
| 6 |
-
"eval_steps": 100,
|
| 7 |
-
"global_step": 1000,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 0.936768149882904,
|
| 14 |
-
"eval_loss": 0.008251233026385307,
|
| 15 |
-
"eval_runtime": 117.4457,
|
| 16 |
-
"eval_samples_per_second": 2267.669,
|
| 17 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9330529793864755,
|
| 18 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6639679670333862,
|
| 19 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9596591982248662,
|
| 20 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.8990018609372358,
|
| 21 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6536919474601746,
|
| 22 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.8488676021429209,
|
| 23 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.8846836847946726,
|
| 24 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.913791126905881,
|
| 25 |
-
"eval_steps_per_second": 0.46,
|
| 26 |
-
"step": 100
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 1.8711943793911008,
|
| 30 |
-
"eval_loss": 0.004326523281633854,
|
| 31 |
-
"eval_runtime": 118.308,
|
| 32 |
-
"eval_samples_per_second": 2251.141,
|
| 33 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9683099913640971,
|
| 34 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6799858808517456,
|
| 35 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.987669070948898,
|
| 36 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9520018198362147,
|
| 37 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6799858808517456,
|
| 38 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9284143244509058,
|
| 39 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9445886468795847,
|
| 40 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9595322710076808,
|
| 41 |
-
"eval_steps_per_second": 0.456,
|
| 42 |
-
"step": 200
|
| 43 |
-
},
|
| 44 |
-
{
|
| 45 |
-
"epoch": 2.8056206088992974,
|
| 46 |
-
"eval_loss": 0.002782753435894847,
|
| 47 |
-
"eval_runtime": 117.8399,
|
| 48 |
-
"eval_samples_per_second": 2260.083,
|
| 49 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9790110013892539,
|
| 50 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7040826678276062,
|
| 51 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9935758649482886,
|
| 52 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9680662667809197,
|
| 53 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7029732465744019,
|
| 54 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9524469797852624,
|
| 55 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9648143930767479,
|
| 56 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9713401352745615,
|
| 57 |
-
"eval_steps_per_second": 0.458,
|
| 58 |
-
"step": 300
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 3.740046838407494,
|
| 62 |
-
"eval_loss": 0.0020659712608903646,
|
| 63 |
-
"eval_runtime": 116.8077,
|
| 64 |
-
"eval_samples_per_second": 2280.056,
|
| 65 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9837419742424811,
|
| 66 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7114190459251404,
|
| 67 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9954100421733855,
|
| 68 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.975348704810703,
|
| 69 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6966520547866821,
|
| 70 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.963270232791414,
|
| 71 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9687853426826509,
|
| 72 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9820016049524246,
|
| 73 |
-
"eval_steps_per_second": 0.462,
|
| 74 |
-
"step": 400
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 4.674473067915691,
|
| 78 |
-
"grad_norm": 0.07067500799894333,
|
| 79 |
-
"learning_rate": 2.9402515723270442e-05,
|
| 80 |
-
"loss": 0.0224,
|
| 81 |
-
"step": 500
|
| 82 |
-
},
|
| 83 |
-
{
|
| 84 |
-
"epoch": 4.674473067915691,
|
| 85 |
-
"eval_loss": 0.0016409169184044003,
|
| 86 |
-
"eval_runtime": 117.7739,
|
| 87 |
-
"eval_samples_per_second": 2261.35,
|
| 88 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.986370292494274,
|
| 89 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7391290664672852,
|
| 90 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.996439193909599,
|
| 91 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9792820044518008,
|
| 92 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7391290664672852,
|
| 93 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9691467317957321,
|
| 94 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.975107979086156,
|
| 95 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9834919179181474,
|
| 96 |
-
"eval_steps_per_second": 0.459,
|
| 97 |
-
"step": 500
|
| 98 |
-
},
|
| 99 |
-
{
|
| 100 |
-
"epoch": 5.608899297423887,
|
| 101 |
-
"eval_loss": 0.0014551315689459443,
|
| 102 |
-
"eval_runtime": 117.5801,
|
| 103 |
-
"eval_samples_per_second": 2265.077,
|
| 104 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9884729470957083,
|
| 105 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7460525035858154,
|
| 106 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9969945004512654,
|
| 107 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9824360661365067,
|
| 108 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7435637712478638,
|
| 109 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9738614226726382,
|
| 110 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9805847418912745,
|
| 111 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9842943941304597,
|
| 112 |
-
"eval_steps_per_second": 0.459,
|
| 113 |
-
"step": 600
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"epoch": 6.543325526932084,
|
| 117 |
-
"eval_loss": 0.0013776659034192562,
|
| 118 |
-
"eval_runtime": 117.6764,
|
| 119 |
-
"eval_samples_per_second": 2263.223,
|
| 120 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9893740847820374,
|
| 121 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7209540009498596,
|
| 122 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.997357375070481,
|
| 123 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9838035826704058,
|
| 124 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7209540009498596,
|
| 125 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9758996171607873,
|
| 126 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9822857142857143,
|
| 127 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9853261492605755,
|
| 128 |
-
"eval_steps_per_second": 0.459,
|
| 129 |
-
"step": 700
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 7.477751756440281,
|
| 133 |
-
"eval_loss": 0.0013444514479488134,
|
| 134 |
-
"eval_runtime": 117.3408,
|
| 135 |
-
"eval_samples_per_second": 2269.696,
|
| 136 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9898246536252018,
|
| 137 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7261425852775574,
|
| 138 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9975494130839752,
|
| 139 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9844654628833477,
|
| 140 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7227741479873657,
|
| 141 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9769000718683564,
|
| 142 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9845218986470993,
|
| 143 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9844090335893615,
|
| 144 |
-
"eval_steps_per_second": 0.46,
|
| 145 |
-
"step": 800
|
| 146 |
-
},
|
| 147 |
-
{
|
| 148 |
-
"epoch": 8.412177985948478,
|
| 149 |
-
"eval_loss": 0.0012511691311374307,
|
| 150 |
-
"eval_runtime": 117.668,
|
| 151 |
-
"eval_samples_per_second": 2263.385,
|
| 152 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9902752224683663,
|
| 153 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.685534656047821,
|
| 154 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9977460917001926,
|
| 155 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9852413242919824,
|
| 156 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6582455635070801,
|
| 157 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9780277137066985,
|
| 158 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9794924087922049,
|
| 159 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9910581222056631,
|
| 160 |
-
"eval_steps_per_second": 0.459,
|
| 161 |
-
"step": 900
|
| 162 |
-
},
|
| 163 |
-
{
|
| 164 |
-
"epoch": 9.346604215456674,
|
| 165 |
-
"grad_norm": 0.018028028309345245,
|
| 166 |
-
"learning_rate": 3.1970649895178203e-06,
|
| 167 |
-
"loss": 0.0052,
|
| 168 |
-
"step": 1000
|
| 169 |
-
},
|
| 170 |
-
{
|
| 171 |
-
"epoch": 9.346604215456674,
|
| 172 |
-
"eval_loss": 0.0012360884575173259,
|
| 173 |
-
"eval_runtime": 117.4598,
|
| 174 |
-
"eval_samples_per_second": 2267.396,
|
| 175 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9905380542935456,
|
| 176 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6790644526481628,
|
| 177 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9977983578816215,
|
| 178 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9856131536880567,
|
| 179 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6790644526481628,
|
| 180 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9785817179348335,
|
| 181 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9816899806664392,
|
| 182 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9895678092399404,
|
| 183 |
-
"eval_steps_per_second": 0.46,
|
| 184 |
-
"step": 1000
|
| 185 |
-
}
|
| 186 |
-
],
|
| 187 |
-
"logging_steps": 500,
|
| 188 |
-
"max_steps": 1060,
|
| 189 |
-
"num_input_tokens_seen": 0,
|
| 190 |
-
"num_train_epochs": 10,
|
| 191 |
-
"save_steps": 100,
|
| 192 |
-
"stateful_callbacks": {
|
| 193 |
-
"EarlyStoppingCallback": {
|
| 194 |
-
"args": {
|
| 195 |
-
"early_stopping_patience": 2,
|
| 196 |
-
"early_stopping_threshold": 0.0
|
| 197 |
-
},
|
| 198 |
-
"attributes": {
|
| 199 |
-
"early_stopping_patience_counter": 0
|
| 200 |
-
}
|
| 201 |
-
},
|
| 202 |
-
"TrainerControl": {
|
| 203 |
-
"args": {
|
| 204 |
-
"should_epoch_stop": false,
|
| 205 |
-
"should_evaluate": false,
|
| 206 |
-
"should_log": false,
|
| 207 |
-
"should_save": true,
|
| 208 |
-
"should_training_stop": false
|
| 209 |
-
},
|
| 210 |
-
"attributes": {}
|
| 211 |
-
}
|
| 212 |
-
},
|
| 213 |
-
"total_flos": 0.0,
|
| 214 |
-
"train_batch_size": 5000,
|
| 215 |
-
"trial_name": null,
|
| 216 |
-
"trial_params": null
|
| 217 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:319aaa354e613c6db82c6bb78290f3da04198ef2c7a75b61b314fa305ed33c45
|
| 3 |
-
size 6033
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1000/unigram.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:da145b5e7700ae40f16691ec32a0b1fdc1ee3298db22a31ea55f57a966c4a65d
|
| 3 |
-
size 14763260
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/1_Pooling/config.json
DELETED
|
@@ -1,10 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"word_embedding_dimension": 384,
|
| 3 |
-
"pooling_mode_cls_token": false,
|
| 4 |
-
"pooling_mode_mean_tokens": true,
|
| 5 |
-
"pooling_mode_max_tokens": false,
|
| 6 |
-
"pooling_mode_mean_sqrt_len_tokens": false,
|
| 7 |
-
"pooling_mode_weightedmean_tokens": false,
|
| 8 |
-
"pooling_mode_lasttoken": false,
|
| 9 |
-
"include_prompt": true
|
| 10 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/README.md
DELETED
|
@@ -1,466 +0,0 @@
|
|
| 1 |
-
---
|
| 2 |
-
language:
|
| 3 |
-
- en
|
| 4 |
-
license: apache-2.0
|
| 5 |
-
tags:
|
| 6 |
-
- sentence-transformers
|
| 7 |
-
- sentence-similarity
|
| 8 |
-
- feature-extraction
|
| 9 |
-
- generated_from_trainer
|
| 10 |
-
- dataset_size:2130620
|
| 11 |
-
- loss:ContrastiveLoss
|
| 12 |
-
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
| 13 |
-
widget:
|
| 14 |
-
- source_sentence: مانوئلا دی سنتا
|
| 15 |
-
sentences:
|
| 16 |
-
- Renko Kitagawa
|
| 17 |
-
- هانس هيرمان وير
|
| 18 |
-
- Ди Чента, Мануэла
|
| 19 |
-
- source_sentence: يورى جافريلوف
|
| 20 |
-
sentences:
|
| 21 |
-
- Wiktor Pinczuk
|
| 22 |
-
- Natalia Germanovna DIRKS
|
| 23 |
-
- Світлана Євгенівна Савицька
|
| 24 |
-
- source_sentence: Џуди Колинс
|
| 25 |
-
sentences:
|
| 26 |
-
- Collins
|
| 27 |
-
- Aisha Muhammed Abdul Salam
|
| 28 |
-
- Phonic Boy On Dope
|
| 29 |
-
- source_sentence: ויליאם בלייר
|
| 30 |
-
sentences:
|
| 31 |
-
- The Hon. Mr Justice Blair
|
| 32 |
-
- Queen Ingrid of Denmark
|
| 33 |
-
- Herman van Rompuy
|
| 34 |
-
- source_sentence: Saif al-Arab GADAFI
|
| 35 |
-
sentences:
|
| 36 |
-
- Максім Недасекаў
|
| 37 |
-
- Mervyn Allister King
|
| 38 |
-
- Paul d. scully-power
|
| 39 |
-
pipeline_tag: sentence-similarity
|
| 40 |
-
library_name: sentence-transformers
|
| 41 |
-
metrics:
|
| 42 |
-
- cosine_accuracy
|
| 43 |
-
- cosine_accuracy_threshold
|
| 44 |
-
- cosine_f1
|
| 45 |
-
- cosine_f1_threshold
|
| 46 |
-
- cosine_precision
|
| 47 |
-
- cosine_recall
|
| 48 |
-
- cosine_ap
|
| 49 |
-
- cosine_mcc
|
| 50 |
-
model-index:
|
| 51 |
-
- name: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2-address-matcher-original
|
| 52 |
-
results:
|
| 53 |
-
- task:
|
| 54 |
-
type: binary-classification
|
| 55 |
-
name: Binary Classification
|
| 56 |
-
dataset:
|
| 57 |
-
name: sentence transformers paraphrase multilingual MiniLM L12 v2
|
| 58 |
-
type: sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2
|
| 59 |
-
metrics:
|
| 60 |
-
- type: cosine_accuracy
|
| 61 |
-
value: 0.9905380542935456
|
| 62 |
-
name: Cosine Accuracy
|
| 63 |
-
- type: cosine_accuracy_threshold
|
| 64 |
-
value: 0.6790644526481628
|
| 65 |
-
name: Cosine Accuracy Threshold
|
| 66 |
-
- type: cosine_f1
|
| 67 |
-
value: 0.9856131536880567
|
| 68 |
-
name: Cosine F1
|
| 69 |
-
- type: cosine_f1_threshold
|
| 70 |
-
value: 0.6790644526481628
|
| 71 |
-
name: Cosine F1 Threshold
|
| 72 |
-
- type: cosine_precision
|
| 73 |
-
value: 0.9816899806664392
|
| 74 |
-
name: Cosine Precision
|
| 75 |
-
- type: cosine_recall
|
| 76 |
-
value: 0.9895678092399404
|
| 77 |
-
name: Cosine Recall
|
| 78 |
-
- type: cosine_ap
|
| 79 |
-
value: 0.9977983578816215
|
| 80 |
-
name: Cosine Ap
|
| 81 |
-
- type: cosine_mcc
|
| 82 |
-
value: 0.9785817179348335
|
| 83 |
-
name: Cosine Mcc
|
| 84 |
-
---
|
| 85 |
-
|
| 86 |
-
# sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2-address-matcher-original
|
| 87 |
-
|
| 88 |
-
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
| 89 |
-
|
| 90 |
-
## Model Details
|
| 91 |
-
|
| 92 |
-
### Model Description
|
| 93 |
-
- **Model Type:** Sentence Transformer
|
| 94 |
-
- **Base model:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) <!-- at revision 86741b4e3f5cb7765a600d3a3d55a0f6a6cb443d -->
|
| 95 |
-
- **Maximum Sequence Length:** 128 tokens
|
| 96 |
-
- **Output Dimensionality:** 384 dimensions
|
| 97 |
-
- **Similarity Function:** Cosine Similarity
|
| 98 |
-
<!-- - **Training Dataset:** Unknown -->
|
| 99 |
-
- **Language:** en
|
| 100 |
-
- **License:** apache-2.0
|
| 101 |
-
|
| 102 |
-
### Model Sources
|
| 103 |
-
|
| 104 |
-
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
| 105 |
-
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
|
| 106 |
-
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
| 107 |
-
|
| 108 |
-
### Full Model Architecture
|
| 109 |
-
|
| 110 |
-
```
|
| 111 |
-
SentenceTransformer(
|
| 112 |
-
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel
|
| 113 |
-
(1): Pooling({'word_embedding_dimension': 384, '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})
|
| 114 |
-
)
|
| 115 |
-
```
|
| 116 |
-
|
| 117 |
-
## Usage
|
| 118 |
-
|
| 119 |
-
### Direct Usage (Sentence Transformers)
|
| 120 |
-
|
| 121 |
-
First install the Sentence Transformers library:
|
| 122 |
-
|
| 123 |
-
```bash
|
| 124 |
-
pip install -U sentence-transformers
|
| 125 |
-
```
|
| 126 |
-
|
| 127 |
-
Then you can load this model and run inference.
|
| 128 |
-
```python
|
| 129 |
-
from sentence_transformers import SentenceTransformer
|
| 130 |
-
|
| 131 |
-
# Download from the 🤗 Hub
|
| 132 |
-
model = SentenceTransformer("sentence_transformers_model_id")
|
| 133 |
-
# Run inference
|
| 134 |
-
sentences = [
|
| 135 |
-
'Saif al-Arab GADAFI',
|
| 136 |
-
'Максім Недасекаў',
|
| 137 |
-
'Mervyn Allister King',
|
| 138 |
-
]
|
| 139 |
-
embeddings = model.encode(sentences)
|
| 140 |
-
print(embeddings.shape)
|
| 141 |
-
# [3, 384]
|
| 142 |
-
|
| 143 |
-
# Get the similarity scores for the embeddings
|
| 144 |
-
similarities = model.similarity(embeddings, embeddings)
|
| 145 |
-
print(similarities.shape)
|
| 146 |
-
# [3, 3]
|
| 147 |
-
```
|
| 148 |
-
|
| 149 |
-
<!--
|
| 150 |
-
### Direct Usage (Transformers)
|
| 151 |
-
|
| 152 |
-
<details><summary>Click to see the direct usage in Transformers</summary>
|
| 153 |
-
|
| 154 |
-
</details>
|
| 155 |
-
-->
|
| 156 |
-
|
| 157 |
-
<!--
|
| 158 |
-
### Downstream Usage (Sentence Transformers)
|
| 159 |
-
|
| 160 |
-
You can finetune this model on your own dataset.
|
| 161 |
-
|
| 162 |
-
<details><summary>Click to expand</summary>
|
| 163 |
-
|
| 164 |
-
</details>
|
| 165 |
-
-->
|
| 166 |
-
|
| 167 |
-
<!--
|
| 168 |
-
### Out-of-Scope Use
|
| 169 |
-
|
| 170 |
-
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
| 171 |
-
-->
|
| 172 |
-
|
| 173 |
-
## Evaluation
|
| 174 |
-
|
| 175 |
-
### Metrics
|
| 176 |
-
|
| 177 |
-
#### Binary Classification
|
| 178 |
-
|
| 179 |
-
* Dataset: `sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2`
|
| 180 |
-
* Evaluated with [<code>BinaryClassificationEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator)
|
| 181 |
-
|
| 182 |
-
| Metric | Value |
|
| 183 |
-
|:--------------------------|:-----------|
|
| 184 |
-
| cosine_accuracy | 0.9905 |
|
| 185 |
-
| cosine_accuracy_threshold | 0.6791 |
|
| 186 |
-
| cosine_f1 | 0.9856 |
|
| 187 |
-
| cosine_f1_threshold | 0.6791 |
|
| 188 |
-
| cosine_precision | 0.9817 |
|
| 189 |
-
| cosine_recall | 0.9896 |
|
| 190 |
-
| **cosine_ap** | **0.9978** |
|
| 191 |
-
| cosine_mcc | 0.9786 |
|
| 192 |
-
|
| 193 |
-
<!--
|
| 194 |
-
## Bias, Risks and Limitations
|
| 195 |
-
|
| 196 |
-
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
| 197 |
-
-->
|
| 198 |
-
|
| 199 |
-
<!--
|
| 200 |
-
### Recommendations
|
| 201 |
-
|
| 202 |
-
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
| 203 |
-
-->
|
| 204 |
-
|
| 205 |
-
## Training Details
|
| 206 |
-
|
| 207 |
-
### Training Dataset
|
| 208 |
-
|
| 209 |
-
#### Unnamed Dataset
|
| 210 |
-
|
| 211 |
-
* Size: 2,130,620 training samples
|
| 212 |
-
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 213 |
-
* Approximate statistics based on the first 1000 samples:
|
| 214 |
-
| | sentence1 | sentence2 | label |
|
| 215 |
-
|:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 216 |
-
| type | string | string | float |
|
| 217 |
-
| details | <ul><li>min: 3 tokens</li><li>mean: 9.28 tokens</li><li>max: 57 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 9.11 tokens</li><li>max: 65 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.34</li><li>max: 1.0</li></ul> |
|
| 218 |
-
* Samples:
|
| 219 |
-
| sentence1 | sentence2 | label |
|
| 220 |
-
|:----------------------------|:-------------------------------|:-----------------|
|
| 221 |
-
| <code>ג'ק וייט</code> | <code>Jack White</code> | <code>1.0</code> |
|
| 222 |
-
| <code>Абдуллоҳ Гул</code> | <code>Савицкая Светлана</code> | <code>0.0</code> |
|
| 223 |
-
| <code>ショーン・ジャスティン・ペン</code> | <code>شان پن</code> | <code>1.0</code> |
|
| 224 |
-
* Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters:
|
| 225 |
-
```json
|
| 226 |
-
{
|
| 227 |
-
"distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
|
| 228 |
-
"margin": 0.5,
|
| 229 |
-
"size_average": true
|
| 230 |
-
}
|
| 231 |
-
```
|
| 232 |
-
|
| 233 |
-
### Evaluation Dataset
|
| 234 |
-
|
| 235 |
-
#### Unnamed Dataset
|
| 236 |
-
|
| 237 |
-
* Size: 266,328 evaluation samples
|
| 238 |
-
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 239 |
-
* Approximate statistics based on the first 1000 samples:
|
| 240 |
-
| | sentence1 | sentence2 | label |
|
| 241 |
-
|:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 242 |
-
| type | string | string | float |
|
| 243 |
-
| details | <ul><li>min: 3 tokens</li><li>mean: 9.27 tokens</li><li>max: 79 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 8.99 tokens</li><li>max: 61 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.32</li><li>max: 1.0</li></ul> |
|
| 244 |
-
* Samples:
|
| 245 |
-
| sentence1 | sentence2 | label |
|
| 246 |
-
|:---------------------------------------------|:-----------------------------------------------|:-----------------|
|
| 247 |
-
| <code>Анатолий Николаевич Герасимов</code> | <code>Anatoli Nikolajewitsch Gerassimow</code> | <code>1.0</code> |
|
| 248 |
-
| <code>Igor Stanislavovitsj Prokopenko</code> | <code>Angelo Lauricella</code> | <code>0.0</code> |
|
| 249 |
-
| <code>Кофе, Линда</code> | <code>Святлана Яўгенаўна Савіцкая</code> | <code>0.0</code> |
|
| 250 |
-
* Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters:
|
| 251 |
-
```json
|
| 252 |
-
{
|
| 253 |
-
"distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
|
| 254 |
-
"margin": 0.5,
|
| 255 |
-
"size_average": true
|
| 256 |
-
}
|
| 257 |
-
```
|
| 258 |
-
|
| 259 |
-
### Training Hyperparameters
|
| 260 |
-
#### Non-Default Hyperparameters
|
| 261 |
-
|
| 262 |
-
- `eval_strategy`: steps
|
| 263 |
-
- `per_device_train_batch_size`: 5000
|
| 264 |
-
- `per_device_eval_batch_size`: 5000
|
| 265 |
-
- `gradient_accumulation_steps`: 4
|
| 266 |
-
- `weight_decay`: 0.02
|
| 267 |
-
- `num_train_epochs`: 10
|
| 268 |
-
- `warmup_ratio`: 0.1
|
| 269 |
-
- `fp16`: True
|
| 270 |
-
- `load_best_model_at_end`: True
|
| 271 |
-
- `optim`: adafactor
|
| 272 |
-
- `gradient_checkpointing`: True
|
| 273 |
-
|
| 274 |
-
#### All Hyperparameters
|
| 275 |
-
<details><summary>Click to expand</summary>
|
| 276 |
-
|
| 277 |
-
- `overwrite_output_dir`: False
|
| 278 |
-
- `do_predict`: False
|
| 279 |
-
- `eval_strategy`: steps
|
| 280 |
-
- `prediction_loss_only`: True
|
| 281 |
-
- `per_device_train_batch_size`: 5000
|
| 282 |
-
- `per_device_eval_batch_size`: 5000
|
| 283 |
-
- `per_gpu_train_batch_size`: None
|
| 284 |
-
- `per_gpu_eval_batch_size`: None
|
| 285 |
-
- `gradient_accumulation_steps`: 4
|
| 286 |
-
- `eval_accumulation_steps`: None
|
| 287 |
-
- `torch_empty_cache_steps`: None
|
| 288 |
-
- `learning_rate`: 5e-05
|
| 289 |
-
- `weight_decay`: 0.02
|
| 290 |
-
- `adam_beta1`: 0.9
|
| 291 |
-
- `adam_beta2`: 0.999
|
| 292 |
-
- `adam_epsilon`: 1e-08
|
| 293 |
-
- `max_grad_norm`: 1.0
|
| 294 |
-
- `num_train_epochs`: 10
|
| 295 |
-
- `max_steps`: -1
|
| 296 |
-
- `lr_scheduler_type`: linear
|
| 297 |
-
- `lr_scheduler_kwargs`: {}
|
| 298 |
-
- `warmup_ratio`: 0.1
|
| 299 |
-
- `warmup_steps`: 0
|
| 300 |
-
- `log_level`: passive
|
| 301 |
-
- `log_level_replica`: warning
|
| 302 |
-
- `log_on_each_node`: True
|
| 303 |
-
- `logging_nan_inf_filter`: True
|
| 304 |
-
- `save_safetensors`: True
|
| 305 |
-
- `save_on_each_node`: False
|
| 306 |
-
- `save_only_model`: False
|
| 307 |
-
- `restore_callback_states_from_checkpoint`: False
|
| 308 |
-
- `no_cuda`: False
|
| 309 |
-
- `use_cpu`: False
|
| 310 |
-
- `use_mps_device`: False
|
| 311 |
-
- `seed`: 42
|
| 312 |
-
- `data_seed`: None
|
| 313 |
-
- `jit_mode_eval`: False
|
| 314 |
-
- `use_ipex`: False
|
| 315 |
-
- `bf16`: False
|
| 316 |
-
- `fp16`: True
|
| 317 |
-
- `fp16_opt_level`: O1
|
| 318 |
-
- `half_precision_backend`: auto
|
| 319 |
-
- `bf16_full_eval`: False
|
| 320 |
-
- `fp16_full_eval`: False
|
| 321 |
-
- `tf32`: None
|
| 322 |
-
- `local_rank`: 0
|
| 323 |
-
- `ddp_backend`: None
|
| 324 |
-
- `tpu_num_cores`: None
|
| 325 |
-
- `tpu_metrics_debug`: False
|
| 326 |
-
- `debug`: []
|
| 327 |
-
- `dataloader_drop_last`: False
|
| 328 |
-
- `dataloader_num_workers`: 0
|
| 329 |
-
- `dataloader_prefetch_factor`: None
|
| 330 |
-
- `past_index`: -1
|
| 331 |
-
- `disable_tqdm`: False
|
| 332 |
-
- `remove_unused_columns`: True
|
| 333 |
-
- `label_names`: None
|
| 334 |
-
- `load_best_model_at_end`: True
|
| 335 |
-
- `ignore_data_skip`: False
|
| 336 |
-
- `fsdp`: []
|
| 337 |
-
- `fsdp_min_num_params`: 0
|
| 338 |
-
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
|
| 339 |
-
- `tp_size`: 0
|
| 340 |
-
- `fsdp_transformer_layer_cls_to_wrap`: None
|
| 341 |
-
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
| 342 |
-
- `deepspeed`: None
|
| 343 |
-
- `label_smoothing_factor`: 0.0
|
| 344 |
-
- `optim`: adafactor
|
| 345 |
-
- `optim_args`: None
|
| 346 |
-
- `adafactor`: False
|
| 347 |
-
- `group_by_length`: False
|
| 348 |
-
- `length_column_name`: length
|
| 349 |
-
- `ddp_find_unused_parameters`: None
|
| 350 |
-
- `ddp_bucket_cap_mb`: None
|
| 351 |
-
- `ddp_broadcast_buffers`: False
|
| 352 |
-
- `dataloader_pin_memory`: True
|
| 353 |
-
- `dataloader_persistent_workers`: False
|
| 354 |
-
- `skip_memory_metrics`: True
|
| 355 |
-
- `use_legacy_prediction_loop`: False
|
| 356 |
-
- `push_to_hub`: False
|
| 357 |
-
- `resume_from_checkpoint`: None
|
| 358 |
-
- `hub_model_id`: None
|
| 359 |
-
- `hub_strategy`: every_save
|
| 360 |
-
- `hub_private_repo`: None
|
| 361 |
-
- `hub_always_push`: False
|
| 362 |
-
- `gradient_checkpointing`: True
|
| 363 |
-
- `gradient_checkpointing_kwargs`: None
|
| 364 |
-
- `include_inputs_for_metrics`: False
|
| 365 |
-
- `include_for_metrics`: []
|
| 366 |
-
- `eval_do_concat_batches`: True
|
| 367 |
-
- `fp16_backend`: auto
|
| 368 |
-
- `push_to_hub_model_id`: None
|
| 369 |
-
- `push_to_hub_organization`: None
|
| 370 |
-
- `mp_parameters`:
|
| 371 |
-
- `auto_find_batch_size`: False
|
| 372 |
-
- `full_determinism`: False
|
| 373 |
-
- `torchdynamo`: None
|
| 374 |
-
- `ray_scope`: last
|
| 375 |
-
- `ddp_timeout`: 1800
|
| 376 |
-
- `torch_compile`: False
|
| 377 |
-
- `torch_compile_backend`: None
|
| 378 |
-
- `torch_compile_mode`: None
|
| 379 |
-
- `include_tokens_per_second`: False
|
| 380 |
-
- `include_num_input_tokens_seen`: False
|
| 381 |
-
- `neftune_noise_alpha`: None
|
| 382 |
-
- `optim_target_modules`: None
|
| 383 |
-
- `batch_eval_metrics`: False
|
| 384 |
-
- `eval_on_start`: False
|
| 385 |
-
- `use_liger_kernel`: False
|
| 386 |
-
- `eval_use_gather_object`: False
|
| 387 |
-
- `average_tokens_across_devices`: False
|
| 388 |
-
- `prompts`: None
|
| 389 |
-
- `batch_sampler`: batch_sampler
|
| 390 |
-
- `multi_dataset_batch_sampler`: proportional
|
| 391 |
-
|
| 392 |
-
</details>
|
| 393 |
-
|
| 394 |
-
### Training Logs
|
| 395 |
-
| Epoch | Step | Training Loss | Validation Loss | sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap |
|
| 396 |
-
|:------:|:----:|:-------------:|:---------------:|:---------------------------------------------------------------------:|
|
| 397 |
-
| -1 | -1 | - | - | 0.7195 |
|
| 398 |
-
| 0.9368 | 100 | - | 0.0083 | 0.9597 |
|
| 399 |
-
| 1.8712 | 200 | - | 0.0043 | 0.9877 |
|
| 400 |
-
| 2.8056 | 300 | - | 0.0028 | 0.9936 |
|
| 401 |
-
| 3.7400 | 400 | - | 0.0021 | 0.9954 |
|
| 402 |
-
| 4.6745 | 500 | 0.0224 | 0.0016 | 0.9964 |
|
| 403 |
-
| 5.6089 | 600 | - | 0.0015 | 0.9970 |
|
| 404 |
-
| 6.5433 | 700 | - | 0.0014 | 0.9974 |
|
| 405 |
-
| 7.4778 | 800 | - | 0.0013 | 0.9975 |
|
| 406 |
-
| 8.4122 | 900 | - | 0.0013 | 0.9977 |
|
| 407 |
-
| 9.3466 | 1000 | 0.0052 | 0.0012 | 0.9978 |
|
| 408 |
-
|
| 409 |
-
|
| 410 |
-
### Framework Versions
|
| 411 |
-
- Python: 3.12.9
|
| 412 |
-
- Sentence Transformers: 3.4.1
|
| 413 |
-
- Transformers: 4.51.3
|
| 414 |
-
- PyTorch: 2.7.0+cu126
|
| 415 |
-
- Accelerate: 1.6.0
|
| 416 |
-
- Datasets: 3.6.0
|
| 417 |
-
- Tokenizers: 0.21.1
|
| 418 |
-
|
| 419 |
-
## Citation
|
| 420 |
-
|
| 421 |
-
### BibTeX
|
| 422 |
-
|
| 423 |
-
#### Sentence Transformers
|
| 424 |
-
```bibtex
|
| 425 |
-
@inproceedings{reimers-2019-sentence-bert,
|
| 426 |
-
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
| 427 |
-
author = "Reimers, Nils and Gurevych, Iryna",
|
| 428 |
-
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
| 429 |
-
month = "11",
|
| 430 |
-
year = "2019",
|
| 431 |
-
publisher = "Association for Computational Linguistics",
|
| 432 |
-
url = "https://arxiv.org/abs/1908.10084",
|
| 433 |
-
}
|
| 434 |
-
```
|
| 435 |
-
|
| 436 |
-
#### ContrastiveLoss
|
| 437 |
-
```bibtex
|
| 438 |
-
@inproceedings{hadsell2006dimensionality,
|
| 439 |
-
author={Hadsell, R. and Chopra, S. and LeCun, Y.},
|
| 440 |
-
booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
|
| 441 |
-
title={Dimensionality Reduction by Learning an Invariant Mapping},
|
| 442 |
-
year={2006},
|
| 443 |
-
volume={2},
|
| 444 |
-
number={},
|
| 445 |
-
pages={1735-1742},
|
| 446 |
-
doi={10.1109/CVPR.2006.100}
|
| 447 |
-
}
|
| 448 |
-
```
|
| 449 |
-
|
| 450 |
-
<!--
|
| 451 |
-
## Glossary
|
| 452 |
-
|
| 453 |
-
*Clearly define terms in order to be accessible across audiences.*
|
| 454 |
-
-->
|
| 455 |
-
|
| 456 |
-
<!--
|
| 457 |
-
## Model Card Authors
|
| 458 |
-
|
| 459 |
-
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
| 460 |
-
-->
|
| 461 |
-
|
| 462 |
-
<!--
|
| 463 |
-
## Model Card Contact
|
| 464 |
-
|
| 465 |
-
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
| 466 |
-
-->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/config.json
DELETED
|
@@ -1,25 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"BertModel"
|
| 4 |
-
],
|
| 5 |
-
"attention_probs_dropout_prob": 0.1,
|
| 6 |
-
"classifier_dropout": null,
|
| 7 |
-
"gradient_checkpointing": false,
|
| 8 |
-
"hidden_act": "gelu",
|
| 9 |
-
"hidden_dropout_prob": 0.1,
|
| 10 |
-
"hidden_size": 384,
|
| 11 |
-
"initializer_range": 0.02,
|
| 12 |
-
"intermediate_size": 1536,
|
| 13 |
-
"layer_norm_eps": 1e-12,
|
| 14 |
-
"max_position_embeddings": 512,
|
| 15 |
-
"model_type": "bert",
|
| 16 |
-
"num_attention_heads": 12,
|
| 17 |
-
"num_hidden_layers": 12,
|
| 18 |
-
"pad_token_id": 0,
|
| 19 |
-
"position_embedding_type": "absolute",
|
| 20 |
-
"torch_dtype": "float32",
|
| 21 |
-
"transformers_version": "4.51.3",
|
| 22 |
-
"type_vocab_size": 2,
|
| 23 |
-
"use_cache": true,
|
| 24 |
-
"vocab_size": 250037
|
| 25 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/config_sentence_transformers.json
DELETED
|
@@ -1,10 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"__version__": {
|
| 3 |
-
"sentence_transformers": "3.4.1",
|
| 4 |
-
"transformers": "4.51.3",
|
| 5 |
-
"pytorch": "2.7.0+cu126"
|
| 6 |
-
},
|
| 7 |
-
"prompts": {},
|
| 8 |
-
"default_prompt_name": null,
|
| 9 |
-
"similarity_fn_name": "cosine"
|
| 10 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:6d56240a57c4c07d9788a1faa198fd34bca9ea0a1e5a26691b1d009dcae94358
|
| 3 |
-
size 470637416
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/modules.json
DELETED
|
@@ -1,14 +0,0 @@
|
|
| 1 |
-
[
|
| 2 |
-
{
|
| 3 |
-
"idx": 0,
|
| 4 |
-
"name": "0",
|
| 5 |
-
"path": "",
|
| 6 |
-
"type": "sentence_transformers.models.Transformer"
|
| 7 |
-
},
|
| 8 |
-
{
|
| 9 |
-
"idx": 1,
|
| 10 |
-
"name": "1",
|
| 11 |
-
"path": "1_Pooling",
|
| 12 |
-
"type": "sentence_transformers.models.Pooling"
|
| 13 |
-
}
|
| 14 |
-
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/optimizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:9b64d5492cb2e0518735a1a0dc7b3b7826a2b4f5d195b44246fdc70db2a64017
|
| 3 |
-
size 1715019
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:8897bcd95f20279aabd5aac16966d704f565763d9f133ce3e3009c72d02b6438
|
| 3 |
-
size 14645
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/scaler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:d3bfa35dd520299a41189dc520e331d371a8b9b17d9abff7077c34c5e038a3b0
|
| 3 |
-
size 1383
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:d60e33f28b0c6c27be860020c700ca71a97176bb114f84fcae7c353e227c8a2e
|
| 3 |
-
size 1465
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/sentence_bert_config.json
DELETED
|
@@ -1,4 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"max_seq_length": 128,
|
| 3 |
-
"do_lower_case": false
|
| 4 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/special_tokens_map.json
DELETED
|
@@ -1,51 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"bos_token": {
|
| 3 |
-
"content": "<s>",
|
| 4 |
-
"lstrip": false,
|
| 5 |
-
"normalized": false,
|
| 6 |
-
"rstrip": false,
|
| 7 |
-
"single_word": false
|
| 8 |
-
},
|
| 9 |
-
"cls_token": {
|
| 10 |
-
"content": "<s>",
|
| 11 |
-
"lstrip": false,
|
| 12 |
-
"normalized": false,
|
| 13 |
-
"rstrip": false,
|
| 14 |
-
"single_word": false
|
| 15 |
-
},
|
| 16 |
-
"eos_token": {
|
| 17 |
-
"content": "</s>",
|
| 18 |
-
"lstrip": false,
|
| 19 |
-
"normalized": false,
|
| 20 |
-
"rstrip": false,
|
| 21 |
-
"single_word": false
|
| 22 |
-
},
|
| 23 |
-
"mask_token": {
|
| 24 |
-
"content": "<mask>",
|
| 25 |
-
"lstrip": true,
|
| 26 |
-
"normalized": false,
|
| 27 |
-
"rstrip": false,
|
| 28 |
-
"single_word": false
|
| 29 |
-
},
|
| 30 |
-
"pad_token": {
|
| 31 |
-
"content": "<pad>",
|
| 32 |
-
"lstrip": false,
|
| 33 |
-
"normalized": false,
|
| 34 |
-
"rstrip": false,
|
| 35 |
-
"single_word": false
|
| 36 |
-
},
|
| 37 |
-
"sep_token": {
|
| 38 |
-
"content": "</s>",
|
| 39 |
-
"lstrip": false,
|
| 40 |
-
"normalized": false,
|
| 41 |
-
"rstrip": false,
|
| 42 |
-
"single_word": false
|
| 43 |
-
},
|
| 44 |
-
"unk_token": {
|
| 45 |
-
"content": "<unk>",
|
| 46 |
-
"lstrip": false,
|
| 47 |
-
"normalized": false,
|
| 48 |
-
"rstrip": false,
|
| 49 |
-
"single_word": false
|
| 50 |
-
}
|
| 51 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/tokenizer.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:cad551d5600a84242d0973327029452a1e3672ba6313c2a3c3d69c4310e12719
|
| 3 |
-
size 17082987
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/tokenizer_config.json
DELETED
|
@@ -1,65 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"added_tokens_decoder": {
|
| 3 |
-
"0": {
|
| 4 |
-
"content": "<s>",
|
| 5 |
-
"lstrip": false,
|
| 6 |
-
"normalized": false,
|
| 7 |
-
"rstrip": false,
|
| 8 |
-
"single_word": false,
|
| 9 |
-
"special": true
|
| 10 |
-
},
|
| 11 |
-
"1": {
|
| 12 |
-
"content": "<pad>",
|
| 13 |
-
"lstrip": false,
|
| 14 |
-
"normalized": false,
|
| 15 |
-
"rstrip": false,
|
| 16 |
-
"single_word": false,
|
| 17 |
-
"special": true
|
| 18 |
-
},
|
| 19 |
-
"2": {
|
| 20 |
-
"content": "</s>",
|
| 21 |
-
"lstrip": false,
|
| 22 |
-
"normalized": false,
|
| 23 |
-
"rstrip": false,
|
| 24 |
-
"single_word": false,
|
| 25 |
-
"special": true
|
| 26 |
-
},
|
| 27 |
-
"3": {
|
| 28 |
-
"content": "<unk>",
|
| 29 |
-
"lstrip": false,
|
| 30 |
-
"normalized": false,
|
| 31 |
-
"rstrip": false,
|
| 32 |
-
"single_word": false,
|
| 33 |
-
"special": true
|
| 34 |
-
},
|
| 35 |
-
"250001": {
|
| 36 |
-
"content": "<mask>",
|
| 37 |
-
"lstrip": true,
|
| 38 |
-
"normalized": false,
|
| 39 |
-
"rstrip": false,
|
| 40 |
-
"single_word": false,
|
| 41 |
-
"special": true
|
| 42 |
-
}
|
| 43 |
-
},
|
| 44 |
-
"bos_token": "<s>",
|
| 45 |
-
"clean_up_tokenization_spaces": false,
|
| 46 |
-
"cls_token": "<s>",
|
| 47 |
-
"do_lower_case": true,
|
| 48 |
-
"eos_token": "</s>",
|
| 49 |
-
"extra_special_tokens": {},
|
| 50 |
-
"mask_token": "<mask>",
|
| 51 |
-
"max_length": 128,
|
| 52 |
-
"model_max_length": 128,
|
| 53 |
-
"pad_to_multiple_of": null,
|
| 54 |
-
"pad_token": "<pad>",
|
| 55 |
-
"pad_token_type_id": 0,
|
| 56 |
-
"padding_side": "right",
|
| 57 |
-
"sep_token": "</s>",
|
| 58 |
-
"stride": 0,
|
| 59 |
-
"strip_accents": null,
|
| 60 |
-
"tokenize_chinese_chars": true,
|
| 61 |
-
"tokenizer_class": "BertTokenizer",
|
| 62 |
-
"truncation_side": "right",
|
| 63 |
-
"truncation_strategy": "longest_first",
|
| 64 |
-
"unk_token": "<unk>"
|
| 65 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/trainer_state.json
DELETED
|
@@ -1,217 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": 1000,
|
| 3 |
-
"best_metric": 0.0012360884575173259,
|
| 4 |
-
"best_model_checkpoint": "data/fine-tuned-sbert-sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2-original-adafactor/checkpoint-1000",
|
| 5 |
-
"epoch": 9.908665105386417,
|
| 6 |
-
"eval_steps": 100,
|
| 7 |
-
"global_step": 1060,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 0.936768149882904,
|
| 14 |
-
"eval_loss": 0.008251233026385307,
|
| 15 |
-
"eval_runtime": 117.4457,
|
| 16 |
-
"eval_samples_per_second": 2267.669,
|
| 17 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9330529793864755,
|
| 18 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6639679670333862,
|
| 19 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9596591982248662,
|
| 20 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.8990018609372358,
|
| 21 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6536919474601746,
|
| 22 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.8488676021429209,
|
| 23 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.8846836847946726,
|
| 24 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.913791126905881,
|
| 25 |
-
"eval_steps_per_second": 0.46,
|
| 26 |
-
"step": 100
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 1.8711943793911008,
|
| 30 |
-
"eval_loss": 0.004326523281633854,
|
| 31 |
-
"eval_runtime": 118.308,
|
| 32 |
-
"eval_samples_per_second": 2251.141,
|
| 33 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9683099913640971,
|
| 34 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6799858808517456,
|
| 35 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.987669070948898,
|
| 36 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9520018198362147,
|
| 37 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6799858808517456,
|
| 38 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9284143244509058,
|
| 39 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9445886468795847,
|
| 40 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9595322710076808,
|
| 41 |
-
"eval_steps_per_second": 0.456,
|
| 42 |
-
"step": 200
|
| 43 |
-
},
|
| 44 |
-
{
|
| 45 |
-
"epoch": 2.8056206088992974,
|
| 46 |
-
"eval_loss": 0.002782753435894847,
|
| 47 |
-
"eval_runtime": 117.8399,
|
| 48 |
-
"eval_samples_per_second": 2260.083,
|
| 49 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9790110013892539,
|
| 50 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7040826678276062,
|
| 51 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9935758649482886,
|
| 52 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9680662667809197,
|
| 53 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7029732465744019,
|
| 54 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9524469797852624,
|
| 55 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9648143930767479,
|
| 56 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9713401352745615,
|
| 57 |
-
"eval_steps_per_second": 0.458,
|
| 58 |
-
"step": 300
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 3.740046838407494,
|
| 62 |
-
"eval_loss": 0.0020659712608903646,
|
| 63 |
-
"eval_runtime": 116.8077,
|
| 64 |
-
"eval_samples_per_second": 2280.056,
|
| 65 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9837419742424811,
|
| 66 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7114190459251404,
|
| 67 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9954100421733855,
|
| 68 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.975348704810703,
|
| 69 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6966520547866821,
|
| 70 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.963270232791414,
|
| 71 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9687853426826509,
|
| 72 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9820016049524246,
|
| 73 |
-
"eval_steps_per_second": 0.462,
|
| 74 |
-
"step": 400
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 4.674473067915691,
|
| 78 |
-
"grad_norm": 0.07067500799894333,
|
| 79 |
-
"learning_rate": 2.9402515723270442e-05,
|
| 80 |
-
"loss": 0.0224,
|
| 81 |
-
"step": 500
|
| 82 |
-
},
|
| 83 |
-
{
|
| 84 |
-
"epoch": 4.674473067915691,
|
| 85 |
-
"eval_loss": 0.0016409169184044003,
|
| 86 |
-
"eval_runtime": 117.7739,
|
| 87 |
-
"eval_samples_per_second": 2261.35,
|
| 88 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.986370292494274,
|
| 89 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7391290664672852,
|
| 90 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.996439193909599,
|
| 91 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9792820044518008,
|
| 92 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7391290664672852,
|
| 93 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9691467317957321,
|
| 94 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.975107979086156,
|
| 95 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9834919179181474,
|
| 96 |
-
"eval_steps_per_second": 0.459,
|
| 97 |
-
"step": 500
|
| 98 |
-
},
|
| 99 |
-
{
|
| 100 |
-
"epoch": 5.608899297423887,
|
| 101 |
-
"eval_loss": 0.0014551315689459443,
|
| 102 |
-
"eval_runtime": 117.5801,
|
| 103 |
-
"eval_samples_per_second": 2265.077,
|
| 104 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9884729470957083,
|
| 105 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7460525035858154,
|
| 106 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9969945004512654,
|
| 107 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9824360661365067,
|
| 108 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7435637712478638,
|
| 109 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9738614226726382,
|
| 110 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9805847418912745,
|
| 111 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9842943941304597,
|
| 112 |
-
"eval_steps_per_second": 0.459,
|
| 113 |
-
"step": 600
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"epoch": 6.543325526932084,
|
| 117 |
-
"eval_loss": 0.0013776659034192562,
|
| 118 |
-
"eval_runtime": 117.6764,
|
| 119 |
-
"eval_samples_per_second": 2263.223,
|
| 120 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9893740847820374,
|
| 121 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7209540009498596,
|
| 122 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.997357375070481,
|
| 123 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9838035826704058,
|
| 124 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7209540009498596,
|
| 125 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9758996171607873,
|
| 126 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9822857142857143,
|
| 127 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9853261492605755,
|
| 128 |
-
"eval_steps_per_second": 0.459,
|
| 129 |
-
"step": 700
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 7.477751756440281,
|
| 133 |
-
"eval_loss": 0.0013444514479488134,
|
| 134 |
-
"eval_runtime": 117.3408,
|
| 135 |
-
"eval_samples_per_second": 2269.696,
|
| 136 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9898246536252018,
|
| 137 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7261425852775574,
|
| 138 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9975494130839752,
|
| 139 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9844654628833477,
|
| 140 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7227741479873657,
|
| 141 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9769000718683564,
|
| 142 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9845218986470993,
|
| 143 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9844090335893615,
|
| 144 |
-
"eval_steps_per_second": 0.46,
|
| 145 |
-
"step": 800
|
| 146 |
-
},
|
| 147 |
-
{
|
| 148 |
-
"epoch": 8.412177985948478,
|
| 149 |
-
"eval_loss": 0.0012511691311374307,
|
| 150 |
-
"eval_runtime": 117.668,
|
| 151 |
-
"eval_samples_per_second": 2263.385,
|
| 152 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9902752224683663,
|
| 153 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.685534656047821,
|
| 154 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9977460917001926,
|
| 155 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9852413242919824,
|
| 156 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6582455635070801,
|
| 157 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9780277137066985,
|
| 158 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9794924087922049,
|
| 159 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9910581222056631,
|
| 160 |
-
"eval_steps_per_second": 0.459,
|
| 161 |
-
"step": 900
|
| 162 |
-
},
|
| 163 |
-
{
|
| 164 |
-
"epoch": 9.346604215456674,
|
| 165 |
-
"grad_norm": 0.018028028309345245,
|
| 166 |
-
"learning_rate": 3.1970649895178203e-06,
|
| 167 |
-
"loss": 0.0052,
|
| 168 |
-
"step": 1000
|
| 169 |
-
},
|
| 170 |
-
{
|
| 171 |
-
"epoch": 9.346604215456674,
|
| 172 |
-
"eval_loss": 0.0012360884575173259,
|
| 173 |
-
"eval_runtime": 117.4598,
|
| 174 |
-
"eval_samples_per_second": 2267.396,
|
| 175 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9905380542935456,
|
| 176 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6790644526481628,
|
| 177 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9977983578816215,
|
| 178 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9856131536880567,
|
| 179 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6790644526481628,
|
| 180 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9785817179348335,
|
| 181 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9816899806664392,
|
| 182 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9895678092399404,
|
| 183 |
-
"eval_steps_per_second": 0.46,
|
| 184 |
-
"step": 1000
|
| 185 |
-
}
|
| 186 |
-
],
|
| 187 |
-
"logging_steps": 500,
|
| 188 |
-
"max_steps": 1060,
|
| 189 |
-
"num_input_tokens_seen": 0,
|
| 190 |
-
"num_train_epochs": 10,
|
| 191 |
-
"save_steps": 100,
|
| 192 |
-
"stateful_callbacks": {
|
| 193 |
-
"EarlyStoppingCallback": {
|
| 194 |
-
"args": {
|
| 195 |
-
"early_stopping_patience": 2,
|
| 196 |
-
"early_stopping_threshold": 0.0
|
| 197 |
-
},
|
| 198 |
-
"attributes": {
|
| 199 |
-
"early_stopping_patience_counter": 0
|
| 200 |
-
}
|
| 201 |
-
},
|
| 202 |
-
"TrainerControl": {
|
| 203 |
-
"args": {
|
| 204 |
-
"should_epoch_stop": false,
|
| 205 |
-
"should_evaluate": false,
|
| 206 |
-
"should_log": false,
|
| 207 |
-
"should_save": true,
|
| 208 |
-
"should_training_stop": true
|
| 209 |
-
},
|
| 210 |
-
"attributes": {}
|
| 211 |
-
}
|
| 212 |
-
},
|
| 213 |
-
"total_flos": 0.0,
|
| 214 |
-
"train_batch_size": 5000,
|
| 215 |
-
"trial_name": null,
|
| 216 |
-
"trial_params": null
|
| 217 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:319aaa354e613c6db82c6bb78290f3da04198ef2c7a75b61b314fa305ed33c45
|
| 3 |
-
size 6033
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-1060/unigram.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:da145b5e7700ae40f16691ec32a0b1fdc1ee3298db22a31ea55f57a966c4a65d
|
| 3 |
-
size 14763260
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/1_Pooling/config.json
DELETED
|
@@ -1,10 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"word_embedding_dimension": 384,
|
| 3 |
-
"pooling_mode_cls_token": false,
|
| 4 |
-
"pooling_mode_mean_tokens": true,
|
| 5 |
-
"pooling_mode_max_tokens": false,
|
| 6 |
-
"pooling_mode_mean_sqrt_len_tokens": false,
|
| 7 |
-
"pooling_mode_weightedmean_tokens": false,
|
| 8 |
-
"pooling_mode_lasttoken": false,
|
| 9 |
-
"include_prompt": true
|
| 10 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/README.md
DELETED
|
@@ -1,463 +0,0 @@
|
|
| 1 |
-
---
|
| 2 |
-
language:
|
| 3 |
-
- en
|
| 4 |
-
license: apache-2.0
|
| 5 |
-
tags:
|
| 6 |
-
- sentence-transformers
|
| 7 |
-
- sentence-similarity
|
| 8 |
-
- feature-extraction
|
| 9 |
-
- generated_from_trainer
|
| 10 |
-
- dataset_size:2130620
|
| 11 |
-
- loss:ContrastiveLoss
|
| 12 |
-
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
|
| 13 |
-
widget:
|
| 14 |
-
- source_sentence: مانوئلا دی سنتا
|
| 15 |
-
sentences:
|
| 16 |
-
- Renko Kitagawa
|
| 17 |
-
- هانس هيرمان وير
|
| 18 |
-
- Ди Чента, Мануэла
|
| 19 |
-
- source_sentence: يورى جافريلوف
|
| 20 |
-
sentences:
|
| 21 |
-
- Wiktor Pinczuk
|
| 22 |
-
- Natalia Germanovna DIRKS
|
| 23 |
-
- Світлана Євгенівна Савицька
|
| 24 |
-
- source_sentence: Џуди Колинс
|
| 25 |
-
sentences:
|
| 26 |
-
- Collins
|
| 27 |
-
- Aisha Muhammed Abdul Salam
|
| 28 |
-
- Phonic Boy On Dope
|
| 29 |
-
- source_sentence: ויליאם בלייר
|
| 30 |
-
sentences:
|
| 31 |
-
- The Hon. Mr Justice Blair
|
| 32 |
-
- Queen Ingrid of Denmark
|
| 33 |
-
- Herman van Rompuy
|
| 34 |
-
- source_sentence: Saif al-Arab GADAFI
|
| 35 |
-
sentences:
|
| 36 |
-
- Максім Недасекаў
|
| 37 |
-
- Mervyn Allister King
|
| 38 |
-
- Paul d. scully-power
|
| 39 |
-
pipeline_tag: sentence-similarity
|
| 40 |
-
library_name: sentence-transformers
|
| 41 |
-
metrics:
|
| 42 |
-
- cosine_accuracy
|
| 43 |
-
- cosine_accuracy_threshold
|
| 44 |
-
- cosine_f1
|
| 45 |
-
- cosine_f1_threshold
|
| 46 |
-
- cosine_precision
|
| 47 |
-
- cosine_recall
|
| 48 |
-
- cosine_ap
|
| 49 |
-
- cosine_mcc
|
| 50 |
-
model-index:
|
| 51 |
-
- name: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2-address-matcher-original
|
| 52 |
-
results:
|
| 53 |
-
- task:
|
| 54 |
-
type: binary-classification
|
| 55 |
-
name: Binary Classification
|
| 56 |
-
dataset:
|
| 57 |
-
name: sentence transformers paraphrase multilingual MiniLM L12 v2
|
| 58 |
-
type: sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2
|
| 59 |
-
metrics:
|
| 60 |
-
- type: cosine_accuracy
|
| 61 |
-
value: 0.9893740847820374
|
| 62 |
-
name: Cosine Accuracy
|
| 63 |
-
- type: cosine_accuracy_threshold
|
| 64 |
-
value: 0.7209540009498596
|
| 65 |
-
name: Cosine Accuracy Threshold
|
| 66 |
-
- type: cosine_f1
|
| 67 |
-
value: 0.9838035826704058
|
| 68 |
-
name: Cosine F1
|
| 69 |
-
- type: cosine_f1_threshold
|
| 70 |
-
value: 0.7209540009498596
|
| 71 |
-
name: Cosine F1 Threshold
|
| 72 |
-
- type: cosine_precision
|
| 73 |
-
value: 0.9822857142857143
|
| 74 |
-
name: Cosine Precision
|
| 75 |
-
- type: cosine_recall
|
| 76 |
-
value: 0.9853261492605755
|
| 77 |
-
name: Cosine Recall
|
| 78 |
-
- type: cosine_ap
|
| 79 |
-
value: 0.997357375070481
|
| 80 |
-
name: Cosine Ap
|
| 81 |
-
- type: cosine_mcc
|
| 82 |
-
value: 0.9758996171607873
|
| 83 |
-
name: Cosine Mcc
|
| 84 |
-
---
|
| 85 |
-
|
| 86 |
-
# sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2-address-matcher-original
|
| 87 |
-
|
| 88 |
-
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
|
| 89 |
-
|
| 90 |
-
## Model Details
|
| 91 |
-
|
| 92 |
-
### Model Description
|
| 93 |
-
- **Model Type:** Sentence Transformer
|
| 94 |
-
- **Base model:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) <!-- at revision 86741b4e3f5cb7765a600d3a3d55a0f6a6cb443d -->
|
| 95 |
-
- **Maximum Sequence Length:** 128 tokens
|
| 96 |
-
- **Output Dimensionality:** 384 dimensions
|
| 97 |
-
- **Similarity Function:** Cosine Similarity
|
| 98 |
-
<!-- - **Training Dataset:** Unknown -->
|
| 99 |
-
- **Language:** en
|
| 100 |
-
- **License:** apache-2.0
|
| 101 |
-
|
| 102 |
-
### Model Sources
|
| 103 |
-
|
| 104 |
-
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
| 105 |
-
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
|
| 106 |
-
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
| 107 |
-
|
| 108 |
-
### Full Model Architecture
|
| 109 |
-
|
| 110 |
-
```
|
| 111 |
-
SentenceTransformer(
|
| 112 |
-
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel
|
| 113 |
-
(1): Pooling({'word_embedding_dimension': 384, '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})
|
| 114 |
-
)
|
| 115 |
-
```
|
| 116 |
-
|
| 117 |
-
## Usage
|
| 118 |
-
|
| 119 |
-
### Direct Usage (Sentence Transformers)
|
| 120 |
-
|
| 121 |
-
First install the Sentence Transformers library:
|
| 122 |
-
|
| 123 |
-
```bash
|
| 124 |
-
pip install -U sentence-transformers
|
| 125 |
-
```
|
| 126 |
-
|
| 127 |
-
Then you can load this model and run inference.
|
| 128 |
-
```python
|
| 129 |
-
from sentence_transformers import SentenceTransformer
|
| 130 |
-
|
| 131 |
-
# Download from the 🤗 Hub
|
| 132 |
-
model = SentenceTransformer("sentence_transformers_model_id")
|
| 133 |
-
# Run inference
|
| 134 |
-
sentences = [
|
| 135 |
-
'Saif al-Arab GADAFI',
|
| 136 |
-
'Максім Недасекаў',
|
| 137 |
-
'Mervyn Allister King',
|
| 138 |
-
]
|
| 139 |
-
embeddings = model.encode(sentences)
|
| 140 |
-
print(embeddings.shape)
|
| 141 |
-
# [3, 384]
|
| 142 |
-
|
| 143 |
-
# Get the similarity scores for the embeddings
|
| 144 |
-
similarities = model.similarity(embeddings, embeddings)
|
| 145 |
-
print(similarities.shape)
|
| 146 |
-
# [3, 3]
|
| 147 |
-
```
|
| 148 |
-
|
| 149 |
-
<!--
|
| 150 |
-
### Direct Usage (Transformers)
|
| 151 |
-
|
| 152 |
-
<details><summary>Click to see the direct usage in Transformers</summary>
|
| 153 |
-
|
| 154 |
-
</details>
|
| 155 |
-
-->
|
| 156 |
-
|
| 157 |
-
<!--
|
| 158 |
-
### Downstream Usage (Sentence Transformers)
|
| 159 |
-
|
| 160 |
-
You can finetune this model on your own dataset.
|
| 161 |
-
|
| 162 |
-
<details><summary>Click to expand</summary>
|
| 163 |
-
|
| 164 |
-
</details>
|
| 165 |
-
-->
|
| 166 |
-
|
| 167 |
-
<!--
|
| 168 |
-
### Out-of-Scope Use
|
| 169 |
-
|
| 170 |
-
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
| 171 |
-
-->
|
| 172 |
-
|
| 173 |
-
## Evaluation
|
| 174 |
-
|
| 175 |
-
### Metrics
|
| 176 |
-
|
| 177 |
-
#### Binary Classification
|
| 178 |
-
|
| 179 |
-
* Dataset: `sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2`
|
| 180 |
-
* Evaluated with [<code>BinaryClassificationEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator)
|
| 181 |
-
|
| 182 |
-
| Metric | Value |
|
| 183 |
-
|:--------------------------|:-----------|
|
| 184 |
-
| cosine_accuracy | 0.9894 |
|
| 185 |
-
| cosine_accuracy_threshold | 0.721 |
|
| 186 |
-
| cosine_f1 | 0.9838 |
|
| 187 |
-
| cosine_f1_threshold | 0.721 |
|
| 188 |
-
| cosine_precision | 0.9823 |
|
| 189 |
-
| cosine_recall | 0.9853 |
|
| 190 |
-
| **cosine_ap** | **0.9974** |
|
| 191 |
-
| cosine_mcc | 0.9759 |
|
| 192 |
-
|
| 193 |
-
<!--
|
| 194 |
-
## Bias, Risks and Limitations
|
| 195 |
-
|
| 196 |
-
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
| 197 |
-
-->
|
| 198 |
-
|
| 199 |
-
<!--
|
| 200 |
-
### Recommendations
|
| 201 |
-
|
| 202 |
-
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
| 203 |
-
-->
|
| 204 |
-
|
| 205 |
-
## Training Details
|
| 206 |
-
|
| 207 |
-
### Training Dataset
|
| 208 |
-
|
| 209 |
-
#### Unnamed Dataset
|
| 210 |
-
|
| 211 |
-
* Size: 2,130,620 training samples
|
| 212 |
-
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 213 |
-
* Approximate statistics based on the first 1000 samples:
|
| 214 |
-
| | sentence1 | sentence2 | label |
|
| 215 |
-
|:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 216 |
-
| type | string | string | float |
|
| 217 |
-
| details | <ul><li>min: 3 tokens</li><li>mean: 9.28 tokens</li><li>max: 57 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 9.11 tokens</li><li>max: 65 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.34</li><li>max: 1.0</li></ul> |
|
| 218 |
-
* Samples:
|
| 219 |
-
| sentence1 | sentence2 | label |
|
| 220 |
-
|:----------------------------|:-------------------------------|:-----------------|
|
| 221 |
-
| <code>ג'ק וייט</code> | <code>Jack White</code> | <code>1.0</code> |
|
| 222 |
-
| <code>Абдуллоҳ Гул</code> | <code>Савицкая Светлана</code> | <code>0.0</code> |
|
| 223 |
-
| <code>ショーン・ジャスティン・ペン</code> | <code>شان پن</code> | <code>1.0</code> |
|
| 224 |
-
* Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters:
|
| 225 |
-
```json
|
| 226 |
-
{
|
| 227 |
-
"distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
|
| 228 |
-
"margin": 0.5,
|
| 229 |
-
"size_average": true
|
| 230 |
-
}
|
| 231 |
-
```
|
| 232 |
-
|
| 233 |
-
### Evaluation Dataset
|
| 234 |
-
|
| 235 |
-
#### Unnamed Dataset
|
| 236 |
-
|
| 237 |
-
* Size: 266,328 evaluation samples
|
| 238 |
-
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 239 |
-
* Approximate statistics based on the first 1000 samples:
|
| 240 |
-
| | sentence1 | sentence2 | label |
|
| 241 |
-
|:--------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------|
|
| 242 |
-
| type | string | string | float |
|
| 243 |
-
| details | <ul><li>min: 3 tokens</li><li>mean: 9.27 tokens</li><li>max: 79 tokens</li></ul> | <ul><li>min: 3 tokens</li><li>mean: 8.99 tokens</li><li>max: 61 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.32</li><li>max: 1.0</li></ul> |
|
| 244 |
-
* Samples:
|
| 245 |
-
| sentence1 | sentence2 | label |
|
| 246 |
-
|:---------------------------------------------|:-----------------------------------------------|:-----------------|
|
| 247 |
-
| <code>Анатолий Николаевич Герасимов</code> | <code>Anatoli Nikolajewitsch Gerassimow</code> | <code>1.0</code> |
|
| 248 |
-
| <code>Igor Stanislavovitsj Prokopenko</code> | <code>Angelo Lauricella</code> | <code>0.0</code> |
|
| 249 |
-
| <code>Кофе, Линда</code> | <code>Святлана Яўгенаўна Савіцкая</code> | <code>0.0</code> |
|
| 250 |
-
* Loss: [<code>ContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#contrastiveloss) with these parameters:
|
| 251 |
-
```json
|
| 252 |
-
{
|
| 253 |
-
"distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE",
|
| 254 |
-
"margin": 0.5,
|
| 255 |
-
"size_average": true
|
| 256 |
-
}
|
| 257 |
-
```
|
| 258 |
-
|
| 259 |
-
### Training Hyperparameters
|
| 260 |
-
#### Non-Default Hyperparameters
|
| 261 |
-
|
| 262 |
-
- `eval_strategy`: steps
|
| 263 |
-
- `per_device_train_batch_size`: 5000
|
| 264 |
-
- `per_device_eval_batch_size`: 5000
|
| 265 |
-
- `gradient_accumulation_steps`: 4
|
| 266 |
-
- `weight_decay`: 0.02
|
| 267 |
-
- `num_train_epochs`: 10
|
| 268 |
-
- `warmup_ratio`: 0.1
|
| 269 |
-
- `fp16`: True
|
| 270 |
-
- `load_best_model_at_end`: True
|
| 271 |
-
- `optim`: adafactor
|
| 272 |
-
- `gradient_checkpointing`: True
|
| 273 |
-
|
| 274 |
-
#### All Hyperparameters
|
| 275 |
-
<details><summary>Click to expand</summary>
|
| 276 |
-
|
| 277 |
-
- `overwrite_output_dir`: False
|
| 278 |
-
- `do_predict`: False
|
| 279 |
-
- `eval_strategy`: steps
|
| 280 |
-
- `prediction_loss_only`: True
|
| 281 |
-
- `per_device_train_batch_size`: 5000
|
| 282 |
-
- `per_device_eval_batch_size`: 5000
|
| 283 |
-
- `per_gpu_train_batch_size`: None
|
| 284 |
-
- `per_gpu_eval_batch_size`: None
|
| 285 |
-
- `gradient_accumulation_steps`: 4
|
| 286 |
-
- `eval_accumulation_steps`: None
|
| 287 |
-
- `torch_empty_cache_steps`: None
|
| 288 |
-
- `learning_rate`: 5e-05
|
| 289 |
-
- `weight_decay`: 0.02
|
| 290 |
-
- `adam_beta1`: 0.9
|
| 291 |
-
- `adam_beta2`: 0.999
|
| 292 |
-
- `adam_epsilon`: 1e-08
|
| 293 |
-
- `max_grad_norm`: 1.0
|
| 294 |
-
- `num_train_epochs`: 10
|
| 295 |
-
- `max_steps`: -1
|
| 296 |
-
- `lr_scheduler_type`: linear
|
| 297 |
-
- `lr_scheduler_kwargs`: {}
|
| 298 |
-
- `warmup_ratio`: 0.1
|
| 299 |
-
- `warmup_steps`: 0
|
| 300 |
-
- `log_level`: passive
|
| 301 |
-
- `log_level_replica`: warning
|
| 302 |
-
- `log_on_each_node`: True
|
| 303 |
-
- `logging_nan_inf_filter`: True
|
| 304 |
-
- `save_safetensors`: True
|
| 305 |
-
- `save_on_each_node`: False
|
| 306 |
-
- `save_only_model`: False
|
| 307 |
-
- `restore_callback_states_from_checkpoint`: False
|
| 308 |
-
- `no_cuda`: False
|
| 309 |
-
- `use_cpu`: False
|
| 310 |
-
- `use_mps_device`: False
|
| 311 |
-
- `seed`: 42
|
| 312 |
-
- `data_seed`: None
|
| 313 |
-
- `jit_mode_eval`: False
|
| 314 |
-
- `use_ipex`: False
|
| 315 |
-
- `bf16`: False
|
| 316 |
-
- `fp16`: True
|
| 317 |
-
- `fp16_opt_level`: O1
|
| 318 |
-
- `half_precision_backend`: auto
|
| 319 |
-
- `bf16_full_eval`: False
|
| 320 |
-
- `fp16_full_eval`: False
|
| 321 |
-
- `tf32`: None
|
| 322 |
-
- `local_rank`: 0
|
| 323 |
-
- `ddp_backend`: None
|
| 324 |
-
- `tpu_num_cores`: None
|
| 325 |
-
- `tpu_metrics_debug`: False
|
| 326 |
-
- `debug`: []
|
| 327 |
-
- `dataloader_drop_last`: False
|
| 328 |
-
- `dataloader_num_workers`: 0
|
| 329 |
-
- `dataloader_prefetch_factor`: None
|
| 330 |
-
- `past_index`: -1
|
| 331 |
-
- `disable_tqdm`: False
|
| 332 |
-
- `remove_unused_columns`: True
|
| 333 |
-
- `label_names`: None
|
| 334 |
-
- `load_best_model_at_end`: True
|
| 335 |
-
- `ignore_data_skip`: False
|
| 336 |
-
- `fsdp`: []
|
| 337 |
-
- `fsdp_min_num_params`: 0
|
| 338 |
-
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
|
| 339 |
-
- `tp_size`: 0
|
| 340 |
-
- `fsdp_transformer_layer_cls_to_wrap`: None
|
| 341 |
-
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
| 342 |
-
- `deepspeed`: None
|
| 343 |
-
- `label_smoothing_factor`: 0.0
|
| 344 |
-
- `optim`: adafactor
|
| 345 |
-
- `optim_args`: None
|
| 346 |
-
- `adafactor`: False
|
| 347 |
-
- `group_by_length`: False
|
| 348 |
-
- `length_column_name`: length
|
| 349 |
-
- `ddp_find_unused_parameters`: None
|
| 350 |
-
- `ddp_bucket_cap_mb`: None
|
| 351 |
-
- `ddp_broadcast_buffers`: False
|
| 352 |
-
- `dataloader_pin_memory`: True
|
| 353 |
-
- `dataloader_persistent_workers`: False
|
| 354 |
-
- `skip_memory_metrics`: True
|
| 355 |
-
- `use_legacy_prediction_loop`: False
|
| 356 |
-
- `push_to_hub`: False
|
| 357 |
-
- `resume_from_checkpoint`: None
|
| 358 |
-
- `hub_model_id`: None
|
| 359 |
-
- `hub_strategy`: every_save
|
| 360 |
-
- `hub_private_repo`: None
|
| 361 |
-
- `hub_always_push`: False
|
| 362 |
-
- `gradient_checkpointing`: True
|
| 363 |
-
- `gradient_checkpointing_kwargs`: None
|
| 364 |
-
- `include_inputs_for_metrics`: False
|
| 365 |
-
- `include_for_metrics`: []
|
| 366 |
-
- `eval_do_concat_batches`: True
|
| 367 |
-
- `fp16_backend`: auto
|
| 368 |
-
- `push_to_hub_model_id`: None
|
| 369 |
-
- `push_to_hub_organization`: None
|
| 370 |
-
- `mp_parameters`:
|
| 371 |
-
- `auto_find_batch_size`: False
|
| 372 |
-
- `full_determinism`: False
|
| 373 |
-
- `torchdynamo`: None
|
| 374 |
-
- `ray_scope`: last
|
| 375 |
-
- `ddp_timeout`: 1800
|
| 376 |
-
- `torch_compile`: False
|
| 377 |
-
- `torch_compile_backend`: None
|
| 378 |
-
- `torch_compile_mode`: None
|
| 379 |
-
- `include_tokens_per_second`: False
|
| 380 |
-
- `include_num_input_tokens_seen`: False
|
| 381 |
-
- `neftune_noise_alpha`: None
|
| 382 |
-
- `optim_target_modules`: None
|
| 383 |
-
- `batch_eval_metrics`: False
|
| 384 |
-
- `eval_on_start`: False
|
| 385 |
-
- `use_liger_kernel`: False
|
| 386 |
-
- `eval_use_gather_object`: False
|
| 387 |
-
- `average_tokens_across_devices`: False
|
| 388 |
-
- `prompts`: None
|
| 389 |
-
- `batch_sampler`: batch_sampler
|
| 390 |
-
- `multi_dataset_batch_sampler`: proportional
|
| 391 |
-
|
| 392 |
-
</details>
|
| 393 |
-
|
| 394 |
-
### Training Logs
|
| 395 |
-
| Epoch | Step | Training Loss | Validation Loss | sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap |
|
| 396 |
-
|:------:|:----:|:-------------:|:---------------:|:---------------------------------------------------------------------:|
|
| 397 |
-
| -1 | -1 | - | - | 0.7195 |
|
| 398 |
-
| 0.9368 | 100 | - | 0.0083 | 0.9597 |
|
| 399 |
-
| 1.8712 | 200 | - | 0.0043 | 0.9877 |
|
| 400 |
-
| 2.8056 | 300 | - | 0.0028 | 0.9936 |
|
| 401 |
-
| 3.7400 | 400 | - | 0.0021 | 0.9954 |
|
| 402 |
-
| 4.6745 | 500 | 0.0224 | 0.0016 | 0.9964 |
|
| 403 |
-
| 5.6089 | 600 | - | 0.0015 | 0.9970 |
|
| 404 |
-
| 6.5433 | 700 | - | 0.0014 | 0.9974 |
|
| 405 |
-
|
| 406 |
-
|
| 407 |
-
### Framework Versions
|
| 408 |
-
- Python: 3.12.9
|
| 409 |
-
- Sentence Transformers: 3.4.1
|
| 410 |
-
- Transformers: 4.51.3
|
| 411 |
-
- PyTorch: 2.7.0+cu126
|
| 412 |
-
- Accelerate: 1.6.0
|
| 413 |
-
- Datasets: 3.6.0
|
| 414 |
-
- Tokenizers: 0.21.1
|
| 415 |
-
|
| 416 |
-
## Citation
|
| 417 |
-
|
| 418 |
-
### BibTeX
|
| 419 |
-
|
| 420 |
-
#### Sentence Transformers
|
| 421 |
-
```bibtex
|
| 422 |
-
@inproceedings{reimers-2019-sentence-bert,
|
| 423 |
-
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
| 424 |
-
author = "Reimers, Nils and Gurevych, Iryna",
|
| 425 |
-
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
| 426 |
-
month = "11",
|
| 427 |
-
year = "2019",
|
| 428 |
-
publisher = "Association for Computational Linguistics",
|
| 429 |
-
url = "https://arxiv.org/abs/1908.10084",
|
| 430 |
-
}
|
| 431 |
-
```
|
| 432 |
-
|
| 433 |
-
#### ContrastiveLoss
|
| 434 |
-
```bibtex
|
| 435 |
-
@inproceedings{hadsell2006dimensionality,
|
| 436 |
-
author={Hadsell, R. and Chopra, S. and LeCun, Y.},
|
| 437 |
-
booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
|
| 438 |
-
title={Dimensionality Reduction by Learning an Invariant Mapping},
|
| 439 |
-
year={2006},
|
| 440 |
-
volume={2},
|
| 441 |
-
number={},
|
| 442 |
-
pages={1735-1742},
|
| 443 |
-
doi={10.1109/CVPR.2006.100}
|
| 444 |
-
}
|
| 445 |
-
```
|
| 446 |
-
|
| 447 |
-
<!--
|
| 448 |
-
## Glossary
|
| 449 |
-
|
| 450 |
-
*Clearly define terms in order to be accessible across audiences.*
|
| 451 |
-
-->
|
| 452 |
-
|
| 453 |
-
<!--
|
| 454 |
-
## Model Card Authors
|
| 455 |
-
|
| 456 |
-
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
| 457 |
-
-->
|
| 458 |
-
|
| 459 |
-
<!--
|
| 460 |
-
## Model Card Contact
|
| 461 |
-
|
| 462 |
-
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
| 463 |
-
-->
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/config.json
DELETED
|
@@ -1,25 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"BertModel"
|
| 4 |
-
],
|
| 5 |
-
"attention_probs_dropout_prob": 0.1,
|
| 6 |
-
"classifier_dropout": null,
|
| 7 |
-
"gradient_checkpointing": false,
|
| 8 |
-
"hidden_act": "gelu",
|
| 9 |
-
"hidden_dropout_prob": 0.1,
|
| 10 |
-
"hidden_size": 384,
|
| 11 |
-
"initializer_range": 0.02,
|
| 12 |
-
"intermediate_size": 1536,
|
| 13 |
-
"layer_norm_eps": 1e-12,
|
| 14 |
-
"max_position_embeddings": 512,
|
| 15 |
-
"model_type": "bert",
|
| 16 |
-
"num_attention_heads": 12,
|
| 17 |
-
"num_hidden_layers": 12,
|
| 18 |
-
"pad_token_id": 0,
|
| 19 |
-
"position_embedding_type": "absolute",
|
| 20 |
-
"torch_dtype": "float32",
|
| 21 |
-
"transformers_version": "4.51.3",
|
| 22 |
-
"type_vocab_size": 2,
|
| 23 |
-
"use_cache": true,
|
| 24 |
-
"vocab_size": 250037
|
| 25 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/config_sentence_transformers.json
DELETED
|
@@ -1,10 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"__version__": {
|
| 3 |
-
"sentence_transformers": "3.4.1",
|
| 4 |
-
"transformers": "4.51.3",
|
| 5 |
-
"pytorch": "2.7.0+cu126"
|
| 6 |
-
},
|
| 7 |
-
"prompts": {},
|
| 8 |
-
"default_prompt_name": null,
|
| 9 |
-
"similarity_fn_name": "cosine"
|
| 10 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:000c7957828311658198adf8e84fe33c1a660836e1cb7b256504f04b8cc770aa
|
| 3 |
-
size 470637416
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/modules.json
DELETED
|
@@ -1,14 +0,0 @@
|
|
| 1 |
-
[
|
| 2 |
-
{
|
| 3 |
-
"idx": 0,
|
| 4 |
-
"name": "0",
|
| 5 |
-
"path": "",
|
| 6 |
-
"type": "sentence_transformers.models.Transformer"
|
| 7 |
-
},
|
| 8 |
-
{
|
| 9 |
-
"idx": 1,
|
| 10 |
-
"name": "1",
|
| 11 |
-
"path": "1_Pooling",
|
| 12 |
-
"type": "sentence_transformers.models.Pooling"
|
| 13 |
-
}
|
| 14 |
-
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/optimizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:e4d52b430164992014c645f921eb89cb4f11af746bb4925e980955f54650d62b
|
| 3 |
-
size 1715019
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:2d56fff20bbc2f130ed2e293f289ea71c316a57e902789c67ac719e6a30c1b4e
|
| 3 |
-
size 14645
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/scaler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:850c3d909f8a0af6f9b431fac5a25833ab1658c39f899825e3b347b6af8a490b
|
| 3 |
-
size 1383
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:d4c2cfbe9b0a118af0de30855464c364252cb3147a7ab4ad3d16c608263feebb
|
| 3 |
-
size 1465
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/sentence_bert_config.json
DELETED
|
@@ -1,4 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"max_seq_length": 128,
|
| 3 |
-
"do_lower_case": false
|
| 4 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/special_tokens_map.json
DELETED
|
@@ -1,51 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"bos_token": {
|
| 3 |
-
"content": "<s>",
|
| 4 |
-
"lstrip": false,
|
| 5 |
-
"normalized": false,
|
| 6 |
-
"rstrip": false,
|
| 7 |
-
"single_word": false
|
| 8 |
-
},
|
| 9 |
-
"cls_token": {
|
| 10 |
-
"content": "<s>",
|
| 11 |
-
"lstrip": false,
|
| 12 |
-
"normalized": false,
|
| 13 |
-
"rstrip": false,
|
| 14 |
-
"single_word": false
|
| 15 |
-
},
|
| 16 |
-
"eos_token": {
|
| 17 |
-
"content": "</s>",
|
| 18 |
-
"lstrip": false,
|
| 19 |
-
"normalized": false,
|
| 20 |
-
"rstrip": false,
|
| 21 |
-
"single_word": false
|
| 22 |
-
},
|
| 23 |
-
"mask_token": {
|
| 24 |
-
"content": "<mask>",
|
| 25 |
-
"lstrip": true,
|
| 26 |
-
"normalized": false,
|
| 27 |
-
"rstrip": false,
|
| 28 |
-
"single_word": false
|
| 29 |
-
},
|
| 30 |
-
"pad_token": {
|
| 31 |
-
"content": "<pad>",
|
| 32 |
-
"lstrip": false,
|
| 33 |
-
"normalized": false,
|
| 34 |
-
"rstrip": false,
|
| 35 |
-
"single_word": false
|
| 36 |
-
},
|
| 37 |
-
"sep_token": {
|
| 38 |
-
"content": "</s>",
|
| 39 |
-
"lstrip": false,
|
| 40 |
-
"normalized": false,
|
| 41 |
-
"rstrip": false,
|
| 42 |
-
"single_word": false
|
| 43 |
-
},
|
| 44 |
-
"unk_token": {
|
| 45 |
-
"content": "<unk>",
|
| 46 |
-
"lstrip": false,
|
| 47 |
-
"normalized": false,
|
| 48 |
-
"rstrip": false,
|
| 49 |
-
"single_word": false
|
| 50 |
-
}
|
| 51 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/tokenizer.json
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:cad551d5600a84242d0973327029452a1e3672ba6313c2a3c3d69c4310e12719
|
| 3 |
-
size 17082987
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/tokenizer_config.json
DELETED
|
@@ -1,65 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"added_tokens_decoder": {
|
| 3 |
-
"0": {
|
| 4 |
-
"content": "<s>",
|
| 5 |
-
"lstrip": false,
|
| 6 |
-
"normalized": false,
|
| 7 |
-
"rstrip": false,
|
| 8 |
-
"single_word": false,
|
| 9 |
-
"special": true
|
| 10 |
-
},
|
| 11 |
-
"1": {
|
| 12 |
-
"content": "<pad>",
|
| 13 |
-
"lstrip": false,
|
| 14 |
-
"normalized": false,
|
| 15 |
-
"rstrip": false,
|
| 16 |
-
"single_word": false,
|
| 17 |
-
"special": true
|
| 18 |
-
},
|
| 19 |
-
"2": {
|
| 20 |
-
"content": "</s>",
|
| 21 |
-
"lstrip": false,
|
| 22 |
-
"normalized": false,
|
| 23 |
-
"rstrip": false,
|
| 24 |
-
"single_word": false,
|
| 25 |
-
"special": true
|
| 26 |
-
},
|
| 27 |
-
"3": {
|
| 28 |
-
"content": "<unk>",
|
| 29 |
-
"lstrip": false,
|
| 30 |
-
"normalized": false,
|
| 31 |
-
"rstrip": false,
|
| 32 |
-
"single_word": false,
|
| 33 |
-
"special": true
|
| 34 |
-
},
|
| 35 |
-
"250001": {
|
| 36 |
-
"content": "<mask>",
|
| 37 |
-
"lstrip": true,
|
| 38 |
-
"normalized": false,
|
| 39 |
-
"rstrip": false,
|
| 40 |
-
"single_word": false,
|
| 41 |
-
"special": true
|
| 42 |
-
}
|
| 43 |
-
},
|
| 44 |
-
"bos_token": "<s>",
|
| 45 |
-
"clean_up_tokenization_spaces": false,
|
| 46 |
-
"cls_token": "<s>",
|
| 47 |
-
"do_lower_case": true,
|
| 48 |
-
"eos_token": "</s>",
|
| 49 |
-
"extra_special_tokens": {},
|
| 50 |
-
"mask_token": "<mask>",
|
| 51 |
-
"max_length": 128,
|
| 52 |
-
"model_max_length": 128,
|
| 53 |
-
"pad_to_multiple_of": null,
|
| 54 |
-
"pad_token": "<pad>",
|
| 55 |
-
"pad_token_type_id": 0,
|
| 56 |
-
"padding_side": "right",
|
| 57 |
-
"sep_token": "</s>",
|
| 58 |
-
"stride": 0,
|
| 59 |
-
"strip_accents": null,
|
| 60 |
-
"tokenize_chinese_chars": true,
|
| 61 |
-
"tokenizer_class": "BertTokenizer",
|
| 62 |
-
"truncation_side": "right",
|
| 63 |
-
"truncation_strategy": "longest_first",
|
| 64 |
-
"unk_token": "<unk>"
|
| 65 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/trainer_state.json
DELETED
|
@@ -1,162 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": 700,
|
| 3 |
-
"best_metric": 0.0013776659034192562,
|
| 4 |
-
"best_model_checkpoint": "data/fine-tuned-sbert-sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2-original-adafactor/checkpoint-700",
|
| 5 |
-
"epoch": 6.543325526932084,
|
| 6 |
-
"eval_steps": 100,
|
| 7 |
-
"global_step": 700,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 0.936768149882904,
|
| 14 |
-
"eval_loss": 0.008251233026385307,
|
| 15 |
-
"eval_runtime": 117.4457,
|
| 16 |
-
"eval_samples_per_second": 2267.669,
|
| 17 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9330529793864755,
|
| 18 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6639679670333862,
|
| 19 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9596591982248662,
|
| 20 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.8990018609372358,
|
| 21 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6536919474601746,
|
| 22 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.8488676021429209,
|
| 23 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.8846836847946726,
|
| 24 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.913791126905881,
|
| 25 |
-
"eval_steps_per_second": 0.46,
|
| 26 |
-
"step": 100
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 1.8711943793911008,
|
| 30 |
-
"eval_loss": 0.004326523281633854,
|
| 31 |
-
"eval_runtime": 118.308,
|
| 32 |
-
"eval_samples_per_second": 2251.141,
|
| 33 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9683099913640971,
|
| 34 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.6799858808517456,
|
| 35 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.987669070948898,
|
| 36 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9520018198362147,
|
| 37 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6799858808517456,
|
| 38 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9284143244509058,
|
| 39 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9445886468795847,
|
| 40 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9595322710076808,
|
| 41 |
-
"eval_steps_per_second": 0.456,
|
| 42 |
-
"step": 200
|
| 43 |
-
},
|
| 44 |
-
{
|
| 45 |
-
"epoch": 2.8056206088992974,
|
| 46 |
-
"eval_loss": 0.002782753435894847,
|
| 47 |
-
"eval_runtime": 117.8399,
|
| 48 |
-
"eval_samples_per_second": 2260.083,
|
| 49 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9790110013892539,
|
| 50 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7040826678276062,
|
| 51 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9935758649482886,
|
| 52 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9680662667809197,
|
| 53 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7029732465744019,
|
| 54 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9524469797852624,
|
| 55 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9648143930767479,
|
| 56 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9713401352745615,
|
| 57 |
-
"eval_steps_per_second": 0.458,
|
| 58 |
-
"step": 300
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 3.740046838407494,
|
| 62 |
-
"eval_loss": 0.0020659712608903646,
|
| 63 |
-
"eval_runtime": 116.8077,
|
| 64 |
-
"eval_samples_per_second": 2280.056,
|
| 65 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9837419742424811,
|
| 66 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7114190459251404,
|
| 67 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9954100421733855,
|
| 68 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.975348704810703,
|
| 69 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.6966520547866821,
|
| 70 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.963270232791414,
|
| 71 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9687853426826509,
|
| 72 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9820016049524246,
|
| 73 |
-
"eval_steps_per_second": 0.462,
|
| 74 |
-
"step": 400
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 4.674473067915691,
|
| 78 |
-
"grad_norm": 0.07067500799894333,
|
| 79 |
-
"learning_rate": 2.9402515723270442e-05,
|
| 80 |
-
"loss": 0.0224,
|
| 81 |
-
"step": 500
|
| 82 |
-
},
|
| 83 |
-
{
|
| 84 |
-
"epoch": 4.674473067915691,
|
| 85 |
-
"eval_loss": 0.0016409169184044003,
|
| 86 |
-
"eval_runtime": 117.7739,
|
| 87 |
-
"eval_samples_per_second": 2261.35,
|
| 88 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.986370292494274,
|
| 89 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7391290664672852,
|
| 90 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.996439193909599,
|
| 91 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9792820044518008,
|
| 92 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7391290664672852,
|
| 93 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9691467317957321,
|
| 94 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.975107979086156,
|
| 95 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9834919179181474,
|
| 96 |
-
"eval_steps_per_second": 0.459,
|
| 97 |
-
"step": 500
|
| 98 |
-
},
|
| 99 |
-
{
|
| 100 |
-
"epoch": 5.608899297423887,
|
| 101 |
-
"eval_loss": 0.0014551315689459443,
|
| 102 |
-
"eval_runtime": 117.5801,
|
| 103 |
-
"eval_samples_per_second": 2265.077,
|
| 104 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9884729470957083,
|
| 105 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7460525035858154,
|
| 106 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.9969945004512654,
|
| 107 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9824360661365067,
|
| 108 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7435637712478638,
|
| 109 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9738614226726382,
|
| 110 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9805847418912745,
|
| 111 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9842943941304597,
|
| 112 |
-
"eval_steps_per_second": 0.459,
|
| 113 |
-
"step": 600
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"epoch": 6.543325526932084,
|
| 117 |
-
"eval_loss": 0.0013776659034192562,
|
| 118 |
-
"eval_runtime": 117.6764,
|
| 119 |
-
"eval_samples_per_second": 2263.223,
|
| 120 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy": 0.9893740847820374,
|
| 121 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_accuracy_threshold": 0.7209540009498596,
|
| 122 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_ap": 0.997357375070481,
|
| 123 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1": 0.9838035826704058,
|
| 124 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_f1_threshold": 0.7209540009498596,
|
| 125 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_mcc": 0.9758996171607873,
|
| 126 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_precision": 0.9822857142857143,
|
| 127 |
-
"eval_sentence-transformers-paraphrase-multilingual-MiniLM-L12-v2_cosine_recall": 0.9853261492605755,
|
| 128 |
-
"eval_steps_per_second": 0.459,
|
| 129 |
-
"step": 700
|
| 130 |
-
}
|
| 131 |
-
],
|
| 132 |
-
"logging_steps": 500,
|
| 133 |
-
"max_steps": 1060,
|
| 134 |
-
"num_input_tokens_seen": 0,
|
| 135 |
-
"num_train_epochs": 10,
|
| 136 |
-
"save_steps": 100,
|
| 137 |
-
"stateful_callbacks": {
|
| 138 |
-
"EarlyStoppingCallback": {
|
| 139 |
-
"args": {
|
| 140 |
-
"early_stopping_patience": 2,
|
| 141 |
-
"early_stopping_threshold": 0.0
|
| 142 |
-
},
|
| 143 |
-
"attributes": {
|
| 144 |
-
"early_stopping_patience_counter": 0
|
| 145 |
-
}
|
| 146 |
-
},
|
| 147 |
-
"TrainerControl": {
|
| 148 |
-
"args": {
|
| 149 |
-
"should_epoch_stop": false,
|
| 150 |
-
"should_evaluate": false,
|
| 151 |
-
"should_log": false,
|
| 152 |
-
"should_save": true,
|
| 153 |
-
"should_training_stop": false
|
| 154 |
-
},
|
| 155 |
-
"attributes": {}
|
| 156 |
-
}
|
| 157 |
-
},
|
| 158 |
-
"total_flos": 0.0,
|
| 159 |
-
"train_batch_size": 5000,
|
| 160 |
-
"trial_name": null,
|
| 161 |
-
"trial_params": null
|
| 162 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint-700/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:319aaa354e613c6db82c6bb78290f3da04198ef2c7a75b61b314fa305ed33c45
|
| 3 |
-
size 6033
|
|
|
|
|
|
|
|
|
|
|
|