saiteki-kai commited on
Commit
427ec0a
·
verified ·
1 Parent(s): 822f22b

Model save

Browse files
Files changed (2) hide show
  1. README.md +86 -0
  2. model.safetensors +1 -1
README.md ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: mit
4
+ base_model: microsoft/deberta-v3-base
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: QA-DeBERTa-v3-base-binary
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # QA-DeBERTa-v3-base-binary
18
+
19
+ This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.3289
22
+ - Accuracy: 0.8592
23
+ - Unsafe Precision: 0.8973
24
+ - Unsafe Recall: 0.8435
25
+ - Unsafe F1: 0.8696
26
+ - Unsafe Fpr: 0.1211
27
+ - Unsafe Aucpr: 0.9534
28
+ - Safe Precision: 0.8174
29
+ - Safe Recall: 0.8789
30
+ - Safe F1: 0.8471
31
+ - Safe Fpr: 0.1565
32
+ - Safe Aucpr: 0.9158
33
+
34
+ ## Model description
35
+
36
+ More information needed
37
+
38
+ ## Intended uses & limitations
39
+
40
+ More information needed
41
+
42
+ ## Training and evaluation data
43
+
44
+ More information needed
45
+
46
+ ## Training procedure
47
+
48
+ ### Training hyperparameters
49
+
50
+ The following hyperparameters were used during training:
51
+ - learning_rate: 6e-06
52
+ - train_batch_size: 64
53
+ - eval_batch_size: 64
54
+ - seed: 42
55
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
56
+ - lr_scheduler_type: linear
57
+ - lr_scheduler_warmup_steps: 1000
58
+ - num_epochs: 10
59
+
60
+ ### Training results
61
+
62
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Unsafe Precision | Unsafe Recall | Unsafe F1 | Unsafe Fpr | Unsafe Aucpr | Safe Precision | Safe Recall | Safe F1 | Safe Fpr | Safe Aucpr |
63
+ |:-------------:|:------:|:-----:|:---------------:|:--------:|:----------------:|:-------------:|:---------:|:----------:|:------------:|:--------------:|:-----------:|:-------:|:--------:|:----------:|
64
+ | 0.3224 | 0.2501 | 2114 | 0.3766 | 0.8332 | 0.8908 | 0.7980 | 0.8419 | 0.1227 | 0.9351 | 0.7759 | 0.8773 | 0.8235 | 0.2020 | 0.8803 |
65
+ | 0.3601 | 0.5001 | 4228 | 0.3495 | 0.8441 | 0.8565 | 0.8647 | 0.8606 | 0.1818 | 0.9433 | 0.8282 | 0.8182 | 0.8232 | 0.1353 | 0.8954 |
66
+ | 0.3229 | 0.7502 | 6342 | 0.3409 | 0.8503 | 0.8891 | 0.8351 | 0.8612 | 0.1307 | 0.9464 | 0.8078 | 0.8693 | 0.8374 | 0.1649 | 0.9012 |
67
+ | 0.3597 | 1.0002 | 8456 | 0.3344 | 0.8520 | 0.8695 | 0.8637 | 0.8666 | 0.1627 | 0.9482 | 0.8304 | 0.8373 | 0.8338 | 0.1363 | 0.9060 |
68
+ | 0.3295 | 1.2503 | 10570 | 0.3357 | 0.8541 | 0.8741 | 0.8619 | 0.8680 | 0.1557 | 0.9491 | 0.8298 | 0.8443 | 0.8370 | 0.1381 | 0.9074 |
69
+ | 0.2894 | 1.5004 | 12684 | 0.3492 | 0.8552 | 0.8710 | 0.8683 | 0.8697 | 0.1613 | 0.9500 | 0.8354 | 0.8387 | 0.8371 | 0.1317 | 0.9096 |
70
+ | 0.2902 | 1.7504 | 14798 | 0.3321 | 0.8564 | 0.8678 | 0.8753 | 0.8715 | 0.1673 | 0.9510 | 0.8418 | 0.8327 | 0.8372 | 0.1247 | 0.9123 |
71
+ | 0.3441 | 2.0005 | 16912 | 0.3246 | 0.8597 | 0.8804 | 0.8654 | 0.8728 | 0.1474 | 0.9524 | 0.8347 | 0.8526 | 0.8435 | 0.1346 | 0.9145 |
72
+ | 0.3268 | 2.2505 | 19026 | 0.3310 | 0.8580 | 0.8664 | 0.8805 | 0.8734 | 0.1704 | 0.9516 | 0.8470 | 0.8296 | 0.8382 | 0.1195 | 0.9135 |
73
+ | 0.2752 | 2.5006 | 21140 | 0.3318 | 0.8580 | 0.8647 | 0.8830 | 0.8737 | 0.1733 | 0.9521 | 0.8492 | 0.8267 | 0.8378 | 0.1170 | 0.9149 |
74
+ | 0.2938 | 2.7507 | 23254 | 0.3236 | 0.8600 | 0.8759 | 0.8720 | 0.8739 | 0.1550 | 0.9529 | 0.8403 | 0.8450 | 0.8426 | 0.1280 | 0.9158 |
75
+ | 0.2993 | 3.0007 | 25368 | 0.3229 | 0.8605 | 0.8826 | 0.8642 | 0.8733 | 0.1443 | 0.9532 | 0.8340 | 0.8557 | 0.8447 | 0.1358 | 0.9166 |
76
+ | 0.2973 | 3.2508 | 27482 | 0.3283 | 0.8604 | 0.8831 | 0.8634 | 0.8731 | 0.1433 | 0.9529 | 0.8333 | 0.8567 | 0.8448 | 0.1366 | 0.9155 |
77
+ | 0.2741 | 3.5008 | 29596 | 0.3288 | 0.8600 | 0.8832 | 0.8625 | 0.8727 | 0.1432 | 0.9531 | 0.8325 | 0.8568 | 0.8445 | 0.1375 | 0.9154 |
78
+ | 0.3123 | 3.7509 | 31710 | 0.3289 | 0.8592 | 0.8973 | 0.8435 | 0.8696 | 0.1211 | 0.9534 | 0.8174 | 0.8789 | 0.8471 | 0.1565 | 0.9158 |
79
+
80
+
81
+ ### Framework versions
82
+
83
+ - Transformers 4.57.3
84
+ - Pytorch 2.7.1+cu118
85
+ - Datasets 4.4.1
86
+ - Tokenizers 0.22.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:188d8e9bd9cc217346e86b455ff7abe4422f99a6f60fd4fef38c4a847aec61d2
3
  size 737719272
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33e3435bfd6a59124f1eb0dacfafa60ea77eb8c3eb21be759281f8610eae269c
3
  size 737719272