Add new SentenceTransformer model
Browse files- README.md +291 -123
- model.safetensors +1 -1
README.md
CHANGED
|
@@ -10,38 +10,34 @@ tags:
|
|
| 10 |
- loss:OnlineContrastiveLoss
|
| 11 |
base_model: sentence-transformers/stsb-distilbert-base
|
| 12 |
widget:
|
| 13 |
-
- source_sentence:
|
| 14 |
sentences:
|
| 15 |
-
-
|
| 16 |
-
|
| 17 |
-
-
|
| 18 |
-
|
| 19 |
-
- source_sentence: What are the safety precautions on handling shotguns proposed by
|
| 20 |
-
the NRA in Idaho?
|
| 21 |
sentences:
|
| 22 |
-
- What
|
| 23 |
-
-
|
| 24 |
-
|
| 25 |
-
-
|
|
|
|
|
|
|
| 26 |
sentences:
|
| 27 |
-
-
|
| 28 |
-
-
|
| 29 |
-
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
sentences:
|
| 33 |
-
-
|
| 34 |
-
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
- source_sentence: Which equation in general relativity predicted the existence of
|
| 38 |
-
black-holes ?
|
| 39 |
sentences:
|
| 40 |
-
-
|
| 41 |
-
-
|
| 42 |
-
|
| 43 |
-
- Why do a lot of theists and agnostics confuse mainstream atheistic thought with
|
| 44 |
-
"positive atheism"?
|
| 45 |
datasets:
|
| 46 |
- sentence-transformers/quora-duplicates
|
| 47 |
pipeline_tag: sentence-similarity
|
|
@@ -89,25 +85,25 @@ model-index:
|
|
| 89 |
value: 0.866
|
| 90 |
name: Cosine Accuracy
|
| 91 |
- type: cosine_accuracy_threshold
|
| 92 |
-
value: 0.
|
| 93 |
name: Cosine Accuracy Threshold
|
| 94 |
- type: cosine_f1
|
| 95 |
-
value: 0.
|
| 96 |
name: Cosine F1
|
| 97 |
- type: cosine_f1_threshold
|
| 98 |
-
value: 0.
|
| 99 |
name: Cosine F1 Threshold
|
| 100 |
- type: cosine_precision
|
| 101 |
-
value: 0.
|
| 102 |
name: Cosine Precision
|
| 103 |
- type: cosine_recall
|
| 104 |
-
value: 0.
|
| 105 |
name: Cosine Recall
|
| 106 |
- type: cosine_ap
|
| 107 |
-
value: 0.
|
| 108 |
name: Cosine Ap
|
| 109 |
- type: cosine_mcc
|
| 110 |
-
value: 0.
|
| 111 |
name: Cosine Mcc
|
| 112 |
- task:
|
| 113 |
type: paraphrase-mining
|
|
@@ -117,19 +113,19 @@ model-index:
|
|
| 117 |
type: quora-duplicates-dev
|
| 118 |
metrics:
|
| 119 |
- type: average_precision
|
| 120 |
-
value: 0.
|
| 121 |
name: Average Precision
|
| 122 |
- type: f1
|
| 123 |
-
value: 0.
|
| 124 |
name: F1
|
| 125 |
- type: precision
|
| 126 |
-
value: 0.
|
| 127 |
name: Precision
|
| 128 |
- type: recall
|
| 129 |
-
value: 0.
|
| 130 |
name: Recall
|
| 131 |
- type: threshold
|
| 132 |
-
value: 0.
|
| 133 |
name: Threshold
|
| 134 |
- task:
|
| 135 |
type: information-retrieval
|
|
@@ -139,49 +135,49 @@ model-index:
|
|
| 139 |
type: unknown
|
| 140 |
metrics:
|
| 141 |
- type: cosine_accuracy@1
|
| 142 |
-
value: 0.
|
| 143 |
name: Cosine Accuracy@1
|
| 144 |
- type: cosine_accuracy@3
|
| 145 |
-
value: 0.
|
| 146 |
name: Cosine Accuracy@3
|
| 147 |
- type: cosine_accuracy@5
|
| 148 |
-
value: 0.
|
| 149 |
name: Cosine Accuracy@5
|
| 150 |
- type: cosine_accuracy@10
|
| 151 |
-
value: 0.
|
| 152 |
name: Cosine Accuracy@10
|
| 153 |
- type: cosine_precision@1
|
| 154 |
-
value: 0.
|
| 155 |
name: Cosine Precision@1
|
| 156 |
- type: cosine_precision@3
|
| 157 |
-
value: 0.
|
| 158 |
name: Cosine Precision@3
|
| 159 |
- type: cosine_precision@5
|
| 160 |
-
value: 0.
|
| 161 |
name: Cosine Precision@5
|
| 162 |
- type: cosine_precision@10
|
| 163 |
-
value: 0.
|
| 164 |
name: Cosine Precision@10
|
| 165 |
- type: cosine_recall@1
|
| 166 |
-
value: 0.
|
| 167 |
name: Cosine Recall@1
|
| 168 |
- type: cosine_recall@3
|
| 169 |
-
value: 0.
|
| 170 |
name: Cosine Recall@3
|
| 171 |
- type: cosine_recall@5
|
| 172 |
-
value: 0.
|
| 173 |
name: Cosine Recall@5
|
| 174 |
- type: cosine_recall@10
|
| 175 |
-
value: 0.
|
| 176 |
name: Cosine Recall@10
|
| 177 |
- type: cosine_ndcg@10
|
| 178 |
-
value: 0.
|
| 179 |
name: Cosine Ndcg@10
|
| 180 |
- type: cosine_mrr@10
|
| 181 |
-
value: 0.
|
| 182 |
name: Cosine Mrr@10
|
| 183 |
- type: cosine_map@100
|
| 184 |
-
value: 0.
|
| 185 |
name: Cosine Map@100
|
| 186 |
---
|
| 187 |
|
|
@@ -235,9 +231,9 @@ from sentence_transformers import SentenceTransformer
|
|
| 235 |
model = SentenceTransformer("CalebR84/stsb-distilbert-base-ocl")
|
| 236 |
# Run inference
|
| 237 |
sentences = [
|
| 238 |
-
'
|
| 239 |
-
'
|
| 240 |
-
'
|
| 241 |
]
|
| 242 |
embeddings = model.encode(sentences)
|
| 243 |
print(embeddings.shape)
|
|
@@ -285,29 +281,29 @@ You can finetune this model on your own dataset.
|
|
| 285 |
| Metric | Value |
|
| 286 |
|:--------------------------|:-----------|
|
| 287 |
| cosine_accuracy | 0.866 |
|
| 288 |
-
| cosine_accuracy_threshold | 0.
|
| 289 |
-
| cosine_f1 | 0.
|
| 290 |
-
| cosine_f1_threshold | 0.
|
| 291 |
-
| cosine_precision | 0.
|
| 292 |
-
| cosine_recall | 0.
|
| 293 |
-
| **cosine_ap** | **0.
|
| 294 |
-
| cosine_mcc | 0.
|
| 295 |
|
| 296 |
#### Paraphrase Mining
|
| 297 |
|
| 298 |
* Dataset: `quora-duplicates-dev`
|
| 299 |
* Evaluated with [<code>ParaphraseMiningEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.ParaphraseMiningEvaluator) with these parameters:
|
| 300 |
```json
|
| 301 |
-
{'add_transitive_closure': <function ParaphraseMiningEvaluator.add_transitive_closure at
|
| 302 |
```
|
| 303 |
|
| 304 |
| Metric | Value |
|
| 305 |
|:----------------------|:-----------|
|
| 306 |
-
| **average_precision** | **0.
|
| 307 |
-
| f1 | 0.
|
| 308 |
-
| precision | 0.
|
| 309 |
-
| recall | 0.
|
| 310 |
-
| threshold | 0.
|
| 311 |
|
| 312 |
#### Information Retrieval
|
| 313 |
|
|
@@ -315,21 +311,21 @@ You can finetune this model on your own dataset.
|
|
| 315 |
|
| 316 |
| Metric | Value |
|
| 317 |
|:--------------------|:-----------|
|
| 318 |
-
| cosine_accuracy@1 | 0.
|
| 319 |
-
| cosine_accuracy@3 | 0.
|
| 320 |
-
| cosine_accuracy@5 | 0.
|
| 321 |
-
| cosine_accuracy@10 | 0.
|
| 322 |
-
| cosine_precision@1 | 0.
|
| 323 |
-
| cosine_precision@3 | 0.
|
| 324 |
-
| cosine_precision@5 | 0.
|
| 325 |
-
| cosine_precision@10 | 0.
|
| 326 |
-
| cosine_recall@1 | 0.
|
| 327 |
-
| cosine_recall@3 | 0.
|
| 328 |
-
| cosine_recall@5 | 0.
|
| 329 |
-
| cosine_recall@10 | 0.
|
| 330 |
-
| **cosine_ndcg@10** | **0.
|
| 331 |
-
| cosine_mrr@10 | 0.
|
| 332 |
-
| cosine_map@100 | 0.
|
| 333 |
|
| 334 |
<!--
|
| 335 |
## Bias, Risks and Limitations
|
|
@@ -353,16 +349,16 @@ You can finetune this model on your own dataset.
|
|
| 353 |
* Size: 100,000 training samples
|
| 354 |
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 355 |
* Approximate statistics based on the first 1000 samples:
|
| 356 |
-
| | sentence1 | sentence2
|
| 357 |
-
|
| 358 |
-
| type | string | string
|
| 359 |
-
| details | <ul><li>min:
|
| 360 |
* Samples:
|
| 361 |
-
| sentence1
|
| 362 |
-
|
| 363 |
-
| <code>
|
| 364 |
-
| <code>How can I
|
| 365 |
-
| <code>I
|
| 366 |
* Loss: [<code>OnlineContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss)
|
| 367 |
|
| 368 |
### Evaluation Dataset
|
|
@@ -376,13 +372,13 @@ You can finetune this model on your own dataset.
|
|
| 376 |
| | sentence1 | sentence2 | label |
|
| 377 |
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------|
|
| 378 |
| type | string | string | int |
|
| 379 |
-
| details | <ul><li>min:
|
| 380 |
* Samples:
|
| 381 |
-
| sentence1
|
| 382 |
-
|
| 383 |
-
| <code>
|
| 384 |
-
| <code>
|
| 385 |
-
| <code>
|
| 386 |
* Loss: [<code>OnlineContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss)
|
| 387 |
|
| 388 |
### Training Hyperparameters
|
|
@@ -391,7 +387,7 @@ You can finetune this model on your own dataset.
|
|
| 391 |
- `eval_strategy`: steps
|
| 392 |
- `per_device_train_batch_size`: 64
|
| 393 |
- `per_device_eval_batch_size`: 64
|
| 394 |
-
- `num_train_epochs`:
|
| 395 |
- `warmup_ratio`: 0.1
|
| 396 |
- `fp16`: True
|
| 397 |
- `batch_sampler`: no_duplicates
|
|
@@ -416,7 +412,7 @@ You can finetune this model on your own dataset.
|
|
| 416 |
- `adam_beta2`: 0.999
|
| 417 |
- `adam_epsilon`: 1e-08
|
| 418 |
- `max_grad_norm`: 1.0
|
| 419 |
-
- `num_train_epochs`:
|
| 420 |
- `max_steps`: -1
|
| 421 |
- `lr_scheduler_type`: linear
|
| 422 |
- `lr_scheduler_kwargs`: {}
|
|
@@ -517,28 +513,200 @@ You can finetune this model on your own dataset.
|
|
| 517 |
</details>
|
| 518 |
|
| 519 |
### Training Logs
|
| 520 |
-
|
| 521 |
-
|
| 522 |
-
|
|
| 523 |
-
|
| 524 |
-
| 0
|
| 525 |
-
| 0.
|
| 526 |
-
| 0.
|
| 527 |
-
| 0.
|
| 528 |
-
| 0.
|
| 529 |
-
| 0.
|
| 530 |
-
| 0.
|
| 531 |
-
| 0.
|
| 532 |
-
| 0.
|
| 533 |
-
| 0.
|
| 534 |
-
| 0.
|
| 535 |
-
| 0.
|
| 536 |
-
| 0.
|
| 537 |
-
| 0.
|
| 538 |
-
| 0.
|
| 539 |
-
| 0.
|
| 540 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 541 |
|
|
|
|
| 542 |
|
| 543 |
### Framework Versions
|
| 544 |
- Python: 3.12.9
|
|
|
|
| 10 |
- loss:OnlineContrastiveLoss
|
| 11 |
base_model: sentence-transformers/stsb-distilbert-base
|
| 12 |
widget:
|
| 13 |
+
- source_sentence: Can I retrieve my deleted text messages on my LG phone?
|
| 14 |
sentences:
|
| 15 |
+
- Why do we sleep?
|
| 16 |
+
- How do I recover a deleted text message from my phone without a computer?
|
| 17 |
+
- What are subjects to study in upsc?
|
| 18 |
+
- source_sentence: How can I prepare for IPS?
|
|
|
|
|
|
|
| 19 |
sentences:
|
| 20 |
+
- What should I prepare for ips?
|
| 21 |
+
- I am trying to find a meaning to life, to give a purpose to my life. Is there
|
| 22 |
+
any book that can help me find my answer, or at least give me the tools?
|
| 23 |
+
- What are the health benefits of Turmeric?
|
| 24 |
+
- source_sentence: Which is the best game development laptop for ₹60,000 to ₹70,000
|
| 25 |
+
INR?
|
| 26 |
sentences:
|
| 27 |
+
- Why doesn't Palestine appear on Google Maps as of 2016?
|
| 28 |
+
- Which is the best laptop for game development under ₹70,000 INR?
|
| 29 |
+
- What is meant by judicial review in the context of the Indian Judiciary?
|
| 30 |
+
- source_sentence: Although light beam bouncing between two plates inside a clock
|
| 31 |
+
is often used to explain time dilation, how can other practical cases be explained?
|
| 32 |
sentences:
|
| 33 |
+
- Is Run Ze Cao's falsification of Einstein's relativity valid?
|
| 34 |
+
- If India denies Pakistan water, will Pakistan give up its nuclear weapons?
|
| 35 |
+
- How do I revise class 12 syllabus in 1 month?
|
| 36 |
+
- source_sentence: How can I lose weight quickly? Need serious help.
|
|
|
|
|
|
|
| 37 |
sentences:
|
| 38 |
+
- Which is the best romantic movie?
|
| 39 |
+
- Why are there so many half-built, abandoned buildings in Mexico?
|
| 40 |
+
- How can you lose weight really quick?
|
|
|
|
|
|
|
| 41 |
datasets:
|
| 42 |
- sentence-transformers/quora-duplicates
|
| 43 |
pipeline_tag: sentence-similarity
|
|
|
|
| 85 |
value: 0.866
|
| 86 |
name: Cosine Accuracy
|
| 87 |
- type: cosine_accuracy_threshold
|
| 88 |
+
value: 0.7860240340232849
|
| 89 |
name: Cosine Accuracy Threshold
|
| 90 |
- type: cosine_f1
|
| 91 |
+
value: 0.8320802005012532
|
| 92 |
name: Cosine F1
|
| 93 |
- type: cosine_f1_threshold
|
| 94 |
+
value: 0.7848798036575317
|
| 95 |
name: Cosine F1 Threshold
|
| 96 |
- type: cosine_precision
|
| 97 |
+
value: 0.7811764705882352
|
| 98 |
name: Cosine Precision
|
| 99 |
- type: cosine_recall
|
| 100 |
+
value: 0.8900804289544236
|
| 101 |
name: Cosine Recall
|
| 102 |
- type: cosine_ap
|
| 103 |
+
value: 0.8772887253419398
|
| 104 |
name: Cosine Ap
|
| 105 |
- type: cosine_mcc
|
| 106 |
+
value: 0.7256385093029618
|
| 107 |
name: Cosine Mcc
|
| 108 |
- task:
|
| 109 |
type: paraphrase-mining
|
|
|
|
| 113 |
type: quora-duplicates-dev
|
| 114 |
metrics:
|
| 115 |
- type: average_precision
|
| 116 |
+
value: 0.6392503009812087
|
| 117 |
name: Average Precision
|
| 118 |
- type: f1
|
| 119 |
+
value: 0.6435291762586327
|
| 120 |
name: F1
|
| 121 |
- type: precision
|
| 122 |
+
value: 0.644658344613225
|
| 123 |
name: Precision
|
| 124 |
- type: recall
|
| 125 |
+
value: 0.6424039566368587
|
| 126 |
name: Recall
|
| 127 |
- type: threshold
|
| 128 |
+
value: 0.8726956844329834
|
| 129 |
name: Threshold
|
| 130 |
- task:
|
| 131 |
type: information-retrieval
|
|
|
|
| 135 |
type: unknown
|
| 136 |
metrics:
|
| 137 |
- type: cosine_accuracy@1
|
| 138 |
+
value: 0.9172
|
| 139 |
name: Cosine Accuracy@1
|
| 140 |
- type: cosine_accuracy@3
|
| 141 |
+
value: 0.9588
|
| 142 |
name: Cosine Accuracy@3
|
| 143 |
- type: cosine_accuracy@5
|
| 144 |
+
value: 0.9672
|
| 145 |
name: Cosine Accuracy@5
|
| 146 |
- type: cosine_accuracy@10
|
| 147 |
+
value: 0.9762
|
| 148 |
name: Cosine Accuracy@10
|
| 149 |
- type: cosine_precision@1
|
| 150 |
+
value: 0.9172
|
| 151 |
name: Cosine Precision@1
|
| 152 |
- type: cosine_precision@3
|
| 153 |
+
value: 0.4102
|
| 154 |
name: Cosine Precision@3
|
| 155 |
- type: cosine_precision@5
|
| 156 |
+
value: 0.2644
|
| 157 |
name: Cosine Precision@5
|
| 158 |
- type: cosine_precision@10
|
| 159 |
+
value: 0.14058
|
| 160 |
name: Cosine Precision@10
|
| 161 |
- type: cosine_recall@1
|
| 162 |
+
value: 0.7868590910037675
|
| 163 |
name: Cosine Recall@1
|
| 164 |
- type: cosine_recall@3
|
| 165 |
+
value: 0.91981069059372
|
| 166 |
name: Cosine Recall@3
|
| 167 |
- type: cosine_recall@5
|
| 168 |
+
value: 0.9442488336402158
|
| 169 |
name: Cosine Recall@5
|
| 170 |
- type: cosine_recall@10
|
| 171 |
+
value: 0.9641439212486859
|
| 172 |
name: Cosine Recall@10
|
| 173 |
- type: cosine_ndcg@10
|
| 174 |
+
value: 0.9388257874901692
|
| 175 |
name: Cosine Ndcg@10
|
| 176 |
- type: cosine_mrr@10
|
| 177 |
+
value: 0.9393049206349205
|
| 178 |
name: Cosine Mrr@10
|
| 179 |
- type: cosine_map@100
|
| 180 |
+
value: 0.9258332306777016
|
| 181 |
name: Cosine Map@100
|
| 182 |
---
|
| 183 |
|
|
|
|
| 231 |
model = SentenceTransformer("CalebR84/stsb-distilbert-base-ocl")
|
| 232 |
# Run inference
|
| 233 |
sentences = [
|
| 234 |
+
'How can I lose weight quickly? Need serious help.',
|
| 235 |
+
'How can you lose weight really quick?',
|
| 236 |
+
'Why are there so many half-built, abandoned buildings in Mexico?',
|
| 237 |
]
|
| 238 |
embeddings = model.encode(sentences)
|
| 239 |
print(embeddings.shape)
|
|
|
|
| 281 |
| Metric | Value |
|
| 282 |
|:--------------------------|:-----------|
|
| 283 |
| cosine_accuracy | 0.866 |
|
| 284 |
+
| cosine_accuracy_threshold | 0.786 |
|
| 285 |
+
| cosine_f1 | 0.8321 |
|
| 286 |
+
| cosine_f1_threshold | 0.7849 |
|
| 287 |
+
| cosine_precision | 0.7812 |
|
| 288 |
+
| cosine_recall | 0.8901 |
|
| 289 |
+
| **cosine_ap** | **0.8773** |
|
| 290 |
+
| cosine_mcc | 0.7256 |
|
| 291 |
|
| 292 |
#### Paraphrase Mining
|
| 293 |
|
| 294 |
* Dataset: `quora-duplicates-dev`
|
| 295 |
* Evaluated with [<code>ParaphraseMiningEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.ParaphraseMiningEvaluator) with these parameters:
|
| 296 |
```json
|
| 297 |
+
{'add_transitive_closure': <function ParaphraseMiningEvaluator.add_transitive_closure at 0x00000219B2FE09A0>, 'max_pairs': 500000, 'top_k': 100}
|
| 298 |
```
|
| 299 |
|
| 300 |
| Metric | Value |
|
| 301 |
|:----------------------|:-----------|
|
| 302 |
+
| **average_precision** | **0.6393** |
|
| 303 |
+
| f1 | 0.6435 |
|
| 304 |
+
| precision | 0.6447 |
|
| 305 |
+
| recall | 0.6424 |
|
| 306 |
+
| threshold | 0.8727 |
|
| 307 |
|
| 308 |
#### Information Retrieval
|
| 309 |
|
|
|
|
| 311 |
|
| 312 |
| Metric | Value |
|
| 313 |
|:--------------------|:-----------|
|
| 314 |
+
| cosine_accuracy@1 | 0.9172 |
|
| 315 |
+
| cosine_accuracy@3 | 0.9588 |
|
| 316 |
+
| cosine_accuracy@5 | 0.9672 |
|
| 317 |
+
| cosine_accuracy@10 | 0.9762 |
|
| 318 |
+
| cosine_precision@1 | 0.9172 |
|
| 319 |
+
| cosine_precision@3 | 0.4102 |
|
| 320 |
+
| cosine_precision@5 | 0.2644 |
|
| 321 |
+
| cosine_precision@10 | 0.1406 |
|
| 322 |
+
| cosine_recall@1 | 0.7869 |
|
| 323 |
+
| cosine_recall@3 | 0.9198 |
|
| 324 |
+
| cosine_recall@5 | 0.9442 |
|
| 325 |
+
| cosine_recall@10 | 0.9641 |
|
| 326 |
+
| **cosine_ndcg@10** | **0.9388** |
|
| 327 |
+
| cosine_mrr@10 | 0.9393 |
|
| 328 |
+
| cosine_map@100 | 0.9258 |
|
| 329 |
|
| 330 |
<!--
|
| 331 |
## Bias, Risks and Limitations
|
|
|
|
| 349 |
* Size: 100,000 training samples
|
| 350 |
* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
|
| 351 |
* Approximate statistics based on the first 1000 samples:
|
| 352 |
+
| | sentence1 | sentence2 | label |
|
| 353 |
+
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------|
|
| 354 |
+
| type | string | string | int |
|
| 355 |
+
| details | <ul><li>min: 6 tokens</li><li>mean: 15.56 tokens</li><li>max: 62 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 15.73 tokens</li><li>max: 84 tokens</li></ul> | <ul><li>0: ~63.20%</li><li>1: ~36.80%</li></ul> |
|
| 356 |
* Samples:
|
| 357 |
+
| sentence1 | sentence2 | label |
|
| 358 |
+
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------|:---------------|
|
| 359 |
+
| <code>What are some of the greatest books not adapted into film yet?</code> | <code>What book should be made into a movie?</code> | <code>0</code> |
|
| 360 |
+
| <code>How can I increase my communication skills?</code> | <code>How we improve our communication skills?</code> | <code>1</code> |
|
| 361 |
+
| <code>Heymen I have a note5 it give me this message when a turn it on and shout down (custom pinary are blocked by frp lock) I try odin and kies butnot work?</code> | <code>Setup dubbing studio with very less budget in India?</code> | <code>0</code> |
|
| 362 |
* Loss: [<code>OnlineContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss)
|
| 363 |
|
| 364 |
### Evaluation Dataset
|
|
|
|
| 372 |
| | sentence1 | sentence2 | label |
|
| 373 |
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------|
|
| 374 |
| type | string | string | int |
|
| 375 |
+
| details | <ul><li>min: 3 tokens</li><li>mean: 15.37 tokens</li><li>max: 62 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 15.63 tokens</li><li>max: 78 tokens</li></ul> | <ul><li>0: ~62.70%</li><li>1: ~37.30%</li></ul> |
|
| 376 |
* Samples:
|
| 377 |
+
| sentence1 | sentence2 | label |
|
| 378 |
+
|:------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------|:---------------|
|
| 379 |
+
| <code>Which is the best book to learn data structures and algorithms?</code> | <code>Which book is the best book for algorithm and datastructure?</code> | <code>1</code> |
|
| 380 |
+
| <code>Does modafinil shows up on a drug test? Because my urine smells a lot of medicine?</code> | <code>Can Modafinil come out in a drug test?</code> | <code>0</code> |
|
| 381 |
+
| <code>Does the size of a penis matter?</code> | <code>Does penis size matters for girls?</code> | <code>1</code> |
|
| 382 |
* Loss: [<code>OnlineContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss)
|
| 383 |
|
| 384 |
### Training Hyperparameters
|
|
|
|
| 387 |
- `eval_strategy`: steps
|
| 388 |
- `per_device_train_batch_size`: 64
|
| 389 |
- `per_device_eval_batch_size`: 64
|
| 390 |
+
- `num_train_epochs`: 10
|
| 391 |
- `warmup_ratio`: 0.1
|
| 392 |
- `fp16`: True
|
| 393 |
- `batch_sampler`: no_duplicates
|
|
|
|
| 412 |
- `adam_beta2`: 0.999
|
| 413 |
- `adam_epsilon`: 1e-08
|
| 414 |
- `max_grad_norm`: 1.0
|
| 415 |
+
- `num_train_epochs`: 10
|
| 416 |
- `max_steps`: -1
|
| 417 |
- `lr_scheduler_type`: linear
|
| 418 |
- `lr_scheduler_kwargs`: {}
|
|
|
|
| 513 |
</details>
|
| 514 |
|
| 515 |
### Training Logs
|
| 516 |
+
<details><summary>Click to expand</summary>
|
| 517 |
+
|
| 518 |
+
| Epoch | Step | Training Loss | Validation Loss | quora-duplicates_cosine_ap | quora-duplicates-dev_average_precision | cosine_ndcg@10 |
|
| 519 |
+
|:------:|:-----:|:-------------:|:---------------:|:--------------------------:|:--------------------------------------:|:--------------:|
|
| 520 |
+
| 0 | 0 | - | - | 0.6905 | 0.4200 | 0.9397 |
|
| 521 |
+
| 0.0640 | 100 | 2.6402 | - | - | - | - |
|
| 522 |
+
| 0.1280 | 200 | 2.4398 | - | - | - | - |
|
| 523 |
+
| 0.1599 | 250 | - | 2.4217 | 0.7392 | 0.4765 | 0.9426 |
|
| 524 |
+
| 0.1919 | 300 | 2.2461 | - | - | - | - |
|
| 525 |
+
| 0.2559 | 400 | 2.1433 | - | - | - | - |
|
| 526 |
+
| 0.3199 | 500 | 2.0417 | 2.1120 | 0.7970 | 0.4566 | 0.9429 |
|
| 527 |
+
| 0.3839 | 600 | 2.0441 | - | - | - | - |
|
| 528 |
+
| 0.4479 | 700 | 1.8907 | - | - | - | - |
|
| 529 |
+
| 0.4798 | 750 | - | 2.0011 | 0.8229 | 0.4820 | 0.9468 |
|
| 530 |
+
| 0.5118 | 800 | 1.8985 | - | - | - | - |
|
| 531 |
+
| 0.5758 | 900 | 1.7521 | - | - | - | - |
|
| 532 |
+
| 0.6398 | 1000 | 1.8888 | 1.8010 | 0.8382 | 0.4925 | 0.9425 |
|
| 533 |
+
| 0.7038 | 1100 | 1.8524 | - | - | - | - |
|
| 534 |
+
| 0.7678 | 1200 | 1.6956 | - | - | - | - |
|
| 535 |
+
| 0.7997 | 1250 | - | 1.8004 | 0.8438 | 0.4283 | 0.9336 |
|
| 536 |
+
| 0.8317 | 1300 | 1.7519 | - | - | - | - |
|
| 537 |
+
| 0.8957 | 1400 | 1.7515 | - | - | - | - |
|
| 538 |
+
| 0.9597 | 1500 | 1.7288 | 1.7434 | 0.8352 | 0.5050 | 0.9428 |
|
| 539 |
+
| 1.0237 | 1600 | 1.533 | - | - | - | - |
|
| 540 |
+
| 1.0877 | 1700 | 1.2543 | - | - | - | - |
|
| 541 |
+
| 1.1196 | 1750 | - | 1.7109 | 0.8514 | 0.5299 | 0.9415 |
|
| 542 |
+
| 1.1516 | 1800 | 1.3201 | - | - | - | - |
|
| 543 |
+
| 1.2156 | 1900 | 1.3309 | - | - | - | - |
|
| 544 |
+
| 1.2796 | 2000 | 1.3256 | 1.7111 | 0.8528 | 0.5138 | 0.9393 |
|
| 545 |
+
| 1.3436 | 2100 | 1.2865 | - | - | - | - |
|
| 546 |
+
| 1.4075 | 2200 | 1.2659 | - | - | - | - |
|
| 547 |
+
| 1.4395 | 2250 | - | 1.7974 | 0.8468 | 0.5320 | 0.9390 |
|
| 548 |
+
| 1.4715 | 2300 | 1.2601 | - | - | - | - |
|
| 549 |
+
| 1.5355 | 2400 | 1.3337 | - | - | - | - |
|
| 550 |
+
| 1.5995 | 2500 | 1.3319 | 1.6922 | 0.8575 | 0.5399 | 0.9416 |
|
| 551 |
+
| 1.6635 | 2600 | 1.3232 | - | - | - | - |
|
| 552 |
+
| 1.7274 | 2700 | 1.3684 | - | - | - | - |
|
| 553 |
+
| 1.7594 | 2750 | - | 1.5772 | 0.8581 | 0.5592 | 0.9484 |
|
| 554 |
+
| 1.7914 | 2800 | 1.2706 | - | - | - | - |
|
| 555 |
+
| 1.8554 | 2900 | 1.3186 | - | - | - | - |
|
| 556 |
+
| 1.9194 | 3000 | 1.2336 | 1.5423 | 0.8656 | 0.5749 | 0.9433 |
|
| 557 |
+
| 1.9834 | 3100 | 1.2193 | - | - | - | - |
|
| 558 |
+
| 2.0473 | 3200 | 0.868 | - | - | - | - |
|
| 559 |
+
| 2.0793 | 3250 | - | 1.6575 | 0.8632 | 0.5735 | 0.9395 |
|
| 560 |
+
| 2.1113 | 3300 | 0.6411 | - | - | - | - |
|
| 561 |
+
| 2.1753 | 3400 | 0.7127 | - | - | - | - |
|
| 562 |
+
| 2.2393 | 3500 | 0.7044 | 1.5778 | 0.8718 | 0.5823 | 0.9387 |
|
| 563 |
+
| 2.3033 | 3600 | 0.6299 | - | - | - | - |
|
| 564 |
+
| 2.3672 | 3700 | 0.7162 | - | - | - | - |
|
| 565 |
+
| 2.3992 | 3750 | - | 1.6300 | 0.8595 | 0.5936 | 0.9414 |
|
| 566 |
+
| 2.4312 | 3800 | 0.6642 | - | - | - | - |
|
| 567 |
+
| 2.4952 | 3900 | 0.6902 | - | - | - | - |
|
| 568 |
+
| 2.5592 | 4000 | 0.7959 | 1.6070 | 0.8637 | 0.6006 | 0.9363 |
|
| 569 |
+
| 2.6232 | 4100 | 0.7588 | - | - | - | - |
|
| 570 |
+
| 2.6871 | 4200 | 0.6925 | - | - | - | - |
|
| 571 |
+
| 2.7191 | 4250 | - | 1.6787 | 0.8682 | 0.6006 | 0.9411 |
|
| 572 |
+
| 2.7511 | 4300 | 0.7226 | - | - | - | - |
|
| 573 |
+
| 2.8151 | 4400 | 0.7507 | - | - | - | - |
|
| 574 |
+
| 2.8791 | 4500 | 0.7563 | 1.6040 | 0.8658 | 0.6061 | 0.9416 |
|
| 575 |
+
| 2.9431 | 4600 | 0.7737 | - | - | - | - |
|
| 576 |
+
| 3.0070 | 4700 | 0.6525 | - | - | - | - |
|
| 577 |
+
| 3.0390 | 4750 | - | 1.6782 | 0.8652 | 0.5983 | 0.9401 |
|
| 578 |
+
| 3.0710 | 4800 | 0.3831 | - | - | - | - |
|
| 579 |
+
| 3.1350 | 4900 | 0.297 | - | - | - | - |
|
| 580 |
+
| 3.1990 | 5000 | 0.3725 | 1.7229 | 0.8588 | 0.6175 | 0.9418 |
|
| 581 |
+
| 3.2630 | 5100 | 0.4142 | - | - | - | - |
|
| 582 |
+
| 3.3269 | 5200 | 0.4415 | - | - | - | - |
|
| 583 |
+
| 3.3589 | 5250 | - | 1.6564 | 0.8635 | 0.6026 | 0.9379 |
|
| 584 |
+
| 3.3909 | 5300 | 0.3729 | - | - | - | - |
|
| 585 |
+
| 3.4549 | 5400 | 0.4164 | - | - | - | - |
|
| 586 |
+
| 3.5189 | 5500 | 0.3668 | 1.5964 | 0.8677 | 0.6105 | 0.9358 |
|
| 587 |
+
| 3.5829 | 5600 | 0.4184 | - | - | - | - |
|
| 588 |
+
| 3.6468 | 5700 | 0.4311 | - | - | - | - |
|
| 589 |
+
| 3.6788 | 5750 | - | 1.6523 | 0.8680 | 0.6130 | 0.9365 |
|
| 590 |
+
| 3.7108 | 5800 | 0.4222 | - | - | - | - |
|
| 591 |
+
| 3.7748 | 5900 | 0.4302 | - | - | - | - |
|
| 592 |
+
| 3.8388 | 6000 | 0.428 | 1.6625 | 0.8674 | 0.6163 | 0.9370 |
|
| 593 |
+
| 3.9028 | 6100 | 0.3898 | - | - | - | - |
|
| 594 |
+
| 3.9667 | 6200 | 0.4255 | - | - | - | - |
|
| 595 |
+
| 3.9987 | 6250 | - | 1.6145 | 0.8680 | 0.6118 | 0.9347 |
|
| 596 |
+
| 4.0307 | 6300 | 0.3456 | - | - | - | - |
|
| 597 |
+
| 4.0947 | 6400 | 0.2265 | - | - | - | - |
|
| 598 |
+
| 4.1587 | 6500 | 0.1913 | 1.7208 | 0.8595 | 0.6339 | 0.9433 |
|
| 599 |
+
| 4.2226 | 6600 | 0.2258 | - | - | - | - |
|
| 600 |
+
| 4.2866 | 6700 | 0.2484 | - | - | - | - |
|
| 601 |
+
| 4.3186 | 6750 | - | 1.6286 | 0.8600 | 0.6313 | 0.9394 |
|
| 602 |
+
| 4.3506 | 6800 | 0.1977 | - | - | - | - |
|
| 603 |
+
| 4.4146 | 6900 | 0.2013 | - | - | - | - |
|
| 604 |
+
| 4.4786 | 7000 | 0.2351 | 1.6910 | 0.8651 | 0.6193 | 0.9401 |
|
| 605 |
+
| 4.5425 | 7100 | 0.2356 | - | - | - | - |
|
| 606 |
+
| 4.6065 | 7200 | 0.2542 | - | - | - | - |
|
| 607 |
+
| 4.6385 | 7250 | - | 1.6955 | 0.8643 | 0.6129 | 0.9357 |
|
| 608 |
+
| 4.6705 | 7300 | 0.2592 | - | - | - | - |
|
| 609 |
+
| 4.7345 | 7400 | 0.2585 | - | - | - | - |
|
| 610 |
+
| 4.7985 | 7500 | 0.2375 | 1.7593 | 0.8647 | 0.6143 | 0.9325 |
|
| 611 |
+
| 4.8624 | 7600 | 0.2506 | - | - | - | - |
|
| 612 |
+
| 4.9264 | 7700 | 0.2394 | - | - | - | - |
|
| 613 |
+
| 4.9584 | 7750 | - | 1.6051 | 0.8720 | 0.6213 | 0.9350 |
|
| 614 |
+
| 4.9904 | 7800 | 0.2374 | - | - | - | - |
|
| 615 |
+
| 5.0544 | 7900 | 0.1675 | - | - | - | - |
|
| 616 |
+
| 5.1184 | 8000 | 0.131 | 1.5864 | 0.8673 | 0.6201 | 0.9377 |
|
| 617 |
+
| 5.1823 | 8100 | 0.1308 | - | - | - | - |
|
| 618 |
+
| 5.2463 | 8200 | 0.1483 | - | - | - | - |
|
| 619 |
+
| 5.2783 | 8250 | - | 1.5976 | 0.8698 | 0.6136 | 0.9359 |
|
| 620 |
+
| 5.3103 | 8300 | 0.1413 | - | - | - | - |
|
| 621 |
+
| 5.3743 | 8400 | 0.1392 | - | - | - | - |
|
| 622 |
+
| 5.4383 | 8500 | 0.1464 | 1.5980 | 0.8661 | 0.6267 | 0.9346 |
|
| 623 |
+
| 5.5022 | 8600 | 0.1781 | - | - | - | - |
|
| 624 |
+
| 5.5662 | 8700 | 0.151 | - | - | - | - |
|
| 625 |
+
| 5.5982 | 8750 | - | 1.5343 | 0.8756 | 0.6245 | 0.9352 |
|
| 626 |
+
| 5.6302 | 8800 | 0.1568 | - | - | - | - |
|
| 627 |
+
| 5.6942 | 8900 | 0.1702 | - | - | - | - |
|
| 628 |
+
| 5.7582 | 9000 | 0.1362 | 1.7121 | 0.8675 | 0.6230 | 0.9362 |
|
| 629 |
+
| 5.8221 | 9100 | 0.1371 | - | - | - | - |
|
| 630 |
+
| 5.8861 | 9200 | 0.1381 | - | - | - | - |
|
| 631 |
+
| 5.9181 | 9250 | - | 1.6326 | 0.8671 | 0.6122 | 0.9302 |
|
| 632 |
+
| 5.9501 | 9300 | 0.1691 | - | - | - | - |
|
| 633 |
+
| 6.0141 | 9400 | 0.1701 | - | - | - | - |
|
| 634 |
+
| 6.0781 | 9500 | 0.0935 | 1.5705 | 0.8709 | 0.6066 | 0.9293 |
|
| 635 |
+
| 6.1420 | 9600 | 0.0852 | - | - | - | - |
|
| 636 |
+
| 6.2060 | 9700 | 0.0874 | - | - | - | - |
|
| 637 |
+
| 6.2380 | 9750 | - | 1.5643 | 0.8724 | 0.6061 | 0.9307 |
|
| 638 |
+
| 6.2700 | 9800 | 0.0889 | - | - | - | - |
|
| 639 |
+
| 6.3340 | 9900 | 0.0972 | - | - | - | - |
|
| 640 |
+
| 6.3980 | 10000 | 0.1011 | 1.5622 | 0.8736 | 0.6153 | 0.9328 |
|
| 641 |
+
| 6.4619 | 10100 | 0.0962 | - | - | - | - |
|
| 642 |
+
| 6.5259 | 10200 | 0.1259 | - | - | - | - |
|
| 643 |
+
| 6.5579 | 10250 | - | 1.5406 | 0.8687 | 0.6293 | 0.9373 |
|
| 644 |
+
| 6.5899 | 10300 | 0.0925 | - | - | - | - |
|
| 645 |
+
| 6.6539 | 10400 | 0.1138 | - | - | - | - |
|
| 646 |
+
| 6.7179 | 10500 | 0.0788 | 1.5450 | 0.8658 | 0.6226 | 0.9349 |
|
| 647 |
+
| 6.7818 | 10600 | 0.1112 | - | - | - | - |
|
| 648 |
+
| 6.8458 | 10700 | 0.0922 | - | - | - | - |
|
| 649 |
+
| 6.8778 | 10750 | - | 1.5063 | 0.8736 | 0.6245 | 0.9370 |
|
| 650 |
+
| 6.9098 | 10800 | 0.1173 | - | - | - | - |
|
| 651 |
+
| 6.9738 | 10900 | 0.1141 | - | - | - | - |
|
| 652 |
+
| 7.0377 | 11000 | 0.0637 | 1.5007 | 0.8741 | 0.6270 | 0.9379 |
|
| 653 |
+
| 7.1017 | 11100 | 0.0713 | - | - | - | - |
|
| 654 |
+
| 7.1657 | 11200 | 0.0754 | - | - | - | - |
|
| 655 |
+
| 7.1977 | 11250 | - | 1.5081 | 0.8725 | 0.6273 | 0.9376 |
|
| 656 |
+
| 7.2297 | 11300 | 0.04 | - | - | - | - |
|
| 657 |
+
| 7.2937 | 11400 | 0.0695 | - | - | - | - |
|
| 658 |
+
| 7.3576 | 11500 | 0.034 | 1.5598 | 0.8710 | 0.6179 | 0.9350 |
|
| 659 |
+
| 7.4216 | 11600 | 0.0513 | - | - | - | - |
|
| 660 |
+
| 7.4856 | 11700 | 0.0749 | - | - | - | - |
|
| 661 |
+
| 7.5176 | 11750 | - | 1.6118 | 0.8694 | 0.6264 | 0.9380 |
|
| 662 |
+
| 7.5496 | 11800 | 0.0708 | - | - | - | - |
|
| 663 |
+
| 7.6136 | 11900 | 0.0939 | - | - | - | - |
|
| 664 |
+
| 7.6775 | 12000 | 0.059 | 1.6282 | 0.8708 | 0.6271 | 0.9354 |
|
| 665 |
+
| 7.7415 | 12100 | 0.0847 | - | - | - | - |
|
| 666 |
+
| 7.8055 | 12200 | 0.0521 | - | - | - | - |
|
| 667 |
+
| 7.8375 | 12250 | - | 1.5478 | 0.8683 | 0.6359 | 0.9388 |
|
| 668 |
+
| 7.8695 | 12300 | 0.0394 | - | - | - | - |
|
| 669 |
+
| 7.9335 | 12400 | 0.0619 | - | - | - | - |
|
| 670 |
+
| 7.9974 | 12500 | 0.0593 | 1.5440 | 0.8771 | 0.6387 | 0.9393 |
|
| 671 |
+
| 8.0614 | 12600 | 0.0292 | - | - | - | - |
|
| 672 |
+
| 8.1254 | 12700 | 0.0267 | - | - | - | - |
|
| 673 |
+
| 8.1574 | 12750 | - | 1.5419 | 0.8773 | 0.6290 | 0.9388 |
|
| 674 |
+
| 8.1894 | 12800 | 0.0334 | - | - | - | - |
|
| 675 |
+
| 8.2534 | 12900 | 0.05 | - | - | - | - |
|
| 676 |
+
| 8.3173 | 13000 | 0.0439 | 1.5589 | 0.8740 | 0.6322 | 0.9384 |
|
| 677 |
+
| 8.3813 | 13100 | 0.0409 | - | - | - | - |
|
| 678 |
+
| 8.4453 | 13200 | 0.03 | - | - | - | - |
|
| 679 |
+
| 8.4773 | 13250 | - | 1.5472 | 0.8730 | 0.6347 | 0.9398 |
|
| 680 |
+
| 8.5093 | 13300 | 0.0373 | - | - | - | - |
|
| 681 |
+
| 8.5733 | 13400 | 0.0404 | - | - | - | - |
|
| 682 |
+
| 8.6372 | 13500 | 0.0357 | 1.5332 | 0.8749 | 0.6327 | 0.9404 |
|
| 683 |
+
| 8.7012 | 13600 | 0.023 | - | - | - | - |
|
| 684 |
+
| 8.7652 | 13700 | 0.0256 | - | - | - | - |
|
| 685 |
+
| 8.7972 | 13750 | - | 1.5154 | 0.8781 | 0.6337 | 0.9379 |
|
| 686 |
+
| 8.8292 | 13800 | 0.0563 | - | - | - | - |
|
| 687 |
+
| 8.8932 | 13900 | 0.029 | - | - | - | - |
|
| 688 |
+
| 8.9571 | 14000 | 0.0395 | 1.5503 | 0.8771 | 0.6344 | 0.9390 |
|
| 689 |
+
| 9.0211 | 14100 | 0.0296 | - | - | - | - |
|
| 690 |
+
| 9.0851 | 14200 | 0.0308 | - | - | - | - |
|
| 691 |
+
| 9.1171 | 14250 | - | 1.5385 | 0.8771 | 0.6363 | 0.9391 |
|
| 692 |
+
| 9.1491 | 14300 | 0.035 | - | - | - | - |
|
| 693 |
+
| 9.2131 | 14400 | 0.0217 | - | - | - | - |
|
| 694 |
+
| 9.2770 | 14500 | 0.0192 | 1.5592 | 0.8777 | 0.6373 | 0.9393 |
|
| 695 |
+
| 9.3410 | 14600 | 0.0369 | - | - | - | - |
|
| 696 |
+
| 9.4050 | 14700 | 0.0186 | - | - | - | - |
|
| 697 |
+
| 9.4370 | 14750 | - | 1.5626 | 0.8771 | 0.6368 | 0.9389 |
|
| 698 |
+
| 9.4690 | 14800 | 0.0303 | - | - | - | - |
|
| 699 |
+
| 9.5329 | 14900 | 0.0181 | - | - | - | - |
|
| 700 |
+
| 9.5969 | 15000 | 0.0217 | 1.5466 | 0.8782 | 0.6387 | 0.9390 |
|
| 701 |
+
| 9.6609 | 15100 | 0.0463 | - | - | - | - |
|
| 702 |
+
| 9.7249 | 15200 | 0.0211 | - | - | - | - |
|
| 703 |
+
| 9.7569 | 15250 | - | 1.5440 | 0.8772 | 0.6401 | 0.9395 |
|
| 704 |
+
| 9.7889 | 15300 | 0.0216 | - | - | - | - |
|
| 705 |
+
| 9.8528 | 15400 | 0.0328 | - | - | - | - |
|
| 706 |
+
| 9.9168 | 15500 | 0.0154 | 1.5399 | 0.8773 | 0.6393 | 0.9388 |
|
| 707 |
+
| 9.9808 | 15600 | 0.0263 | - | - | - | - |
|
| 708 |
|
| 709 |
+
</details>
|
| 710 |
|
| 711 |
### Framework Versions
|
| 712 |
- Python: 3.12.9
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 265462608
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca632dc6a8785edfec19926fc59b3ac105f42543a7657f0164574f74e1ead708
|
| 3 |
size 265462608
|