Hugging Face's logo

Heather-Driver
/
distilbert-classn-LinearAlg-finetuned-span-width-3

Transformers
TensorBoard
Safetensors
distilbert
Generated from Trainer
Model card Files
xet
Metrics Community
Heather-Driver commited on
Commit
5c23bad
·
verified ·
1 Parent(s): afcba6a

Classification Training

Browse files
Files changed (1) hide show
  1. README.md +46 -31
README.md CHANGED
@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [dslim/distilbert-NER](https://huggingface.co/dslim/distilbert-NER) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.9310
25
- - Accuracy: 0.8095
26
- - F1: 0.8097
27
- - Precision: 0.8217
28
- - Recall: 0.8095
29
 
30
  ## Model description
31
 
@@ -48,44 +48,59 @@ The following hyperparameters were used during training:
48
  - train_batch_size: 2
49
  - eval_batch_size: 2
50
  - seed: 42
51
- - gradient_accumulation_steps: 4
52
- - total_train_batch_size: 8
53
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
54
  - lr_scheduler_type: linear
55
  - lr_scheduler_warmup_steps: 500
56
- - num_epochs: 30
57
  - mixed_precision_training: Native AMP
58
 
59
  ### Training results
60
 
61
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
62
  |:-------------:|:-------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
63
- | 9.8014 | 1.3562 | 50 | 2.4817 | 0.0794 | 0.0660 | 0.0588 | 0.0794 |
64
- | 9.5418 | 2.7123 | 100 | 2.3902 | 0.1190 | 0.1015 | 0.0904 | 0.1190 |
65
- | 8.824 | 4.0548 | 150 | 2.2084 | 0.2381 | 0.2168 | 0.2066 | 0.2381 |
66
- | 7.7994 | 5.4110 | 200 | 2.0021 | 0.3730 | 0.3678 | 0.3866 | 0.3730 |
67
- | 6.5913 | 6.7671 | 250 | 1.7473 | 0.5079 | 0.5102 | 0.5691 | 0.5079 |
68
- | 5.3512 | 8.1096 | 300 | 1.4931 | 0.6270 | 0.6316 | 0.6586 | 0.6270 |
69
- | 4.059 | 9.4658 | 350 | 1.2817 | 0.6667 | 0.6746 | 0.6984 | 0.6667 |
70
- | 2.7457 | 10.8219 | 400 | 1.0807 | 0.7063 | 0.7110 | 0.7259 | 0.7063 |
71
- | 1.88 | 12.1644 | 450 | 0.9598 | 0.7460 | 0.7495 | 0.7623 | 0.7460 |
72
- | 1.1331 | 13.5205 | 500 | 0.8816 | 0.7857 | 0.7835 | 0.7981 | 0.7857 |
73
- | 0.7162 | 14.8767 | 550 | 0.8317 | 0.7857 | 0.7841 | 0.7904 | 0.7857 |
74
- | 0.3564 | 16.2192 | 600 | 0.8093 | 0.7857 | 0.7843 | 0.7970 | 0.7857 |
75
- | 0.2064 | 17.5753 | 650 | 0.8265 | 0.7937 | 0.7925 | 0.8014 | 0.7937 |
76
- | 0.1326 | 18.9315 | 700 | 0.8605 | 0.7937 | 0.7941 | 0.8123 | 0.7937 |
77
- | 0.0618 | 20.2740 | 750 | 0.8623 | 0.8095 | 0.8102 | 0.8243 | 0.8095 |
78
- | 0.0487 | 21.6301 | 800 | 0.8693 | 0.8095 | 0.8078 | 0.8177 | 0.8095 |
79
- | 0.0322 | 22.9863 | 850 | 0.8948 | 0.8016 | 0.8034 | 0.8202 | 0.8016 |
80
- | 0.0229 | 24.3288 | 900 | 0.9234 | 0.8016 | 0.8032 | 0.8180 | 0.8016 |
81
- | 0.0202 | 25.6849 | 950 | 0.9288 | 0.8016 | 0.7992 | 0.8121 | 0.8016 |
82
- | 0.0163 | 27.0274 | 1000 | 0.9279 | 0.8095 | 0.8090 | 0.8208 | 0.8095 |
83
- | 0.014 | 28.3836 | 1050 | 0.9310 | 0.8095 | 0.8097 | 0.8217 | 0.8095 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
84
 
85
 
86
  ### Framework versions
87
 
88
- - Transformers 4.48.2
89
  - Pytorch 2.5.1+cu124
90
- - Datasets 3.2.0
91
  - Tokenizers 0.21.0
 
21
 
22
  This model is a fine-tuned version of [dslim/distilbert-NER](https://huggingface.co/dslim/distilbert-NER) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 1.0465
25
+ - Accuracy: 0.7222
26
+ - F1: 0.7251
27
+ - Precision: 0.7350
28
+ - Recall: 0.7222
29
 
30
  ## Model description
31
 
 
48
  - train_batch_size: 2
49
  - eval_batch_size: 2
50
  - seed: 42
51
+ - gradient_accumulation_steps: 2
52
+ - total_train_batch_size: 4
53
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
54
  - lr_scheduler_type: linear
55
  - lr_scheduler_warmup_steps: 500
56
+ - num_epochs: 25
57
  - mixed_precision_training: Native AMP
58
 
59
  ### Training results
60
 
61
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
62
  |:-------------:|:-------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
63
+ | 5.0331 | 0.6849 | 50 | 2.4889 | 0.0476 | 0.0310 | 0.0241 | 0.0476 |
64
+ | 4.8321 | 1.3699 | 100 | 2.4703 | 0.0794 | 0.0509 | 0.0396 | 0.0794 |
65
+ | 4.8811 | 2.0548 | 150 | 2.4370 | 0.0873 | 0.0587 | 0.0487 | 0.0873 |
66
+ | 4.8198 | 2.7397 | 200 | 2.4201 | 0.0952 | 0.0662 | 0.0670 | 0.0952 |
67
+ | 4.7571 | 3.4247 | 250 | 2.4151 | 0.1190 | 0.0876 | 0.1070 | 0.1190 |
68
+ | 4.6927 | 4.1096 | 300 | 2.3845 | 0.1270 | 0.1051 | 0.1038 | 0.1270 |
69
+ | 4.607 | 4.7945 | 350 | 2.3643 | 0.1508 | 0.1431 | 0.1732 | 0.1508 |
70
+ | 4.5543 | 5.4795 | 400 | 2.3641 | 0.1508 | 0.1376 | 0.1467 | 0.1508 |
71
+ | 4.2468 | 6.1644 | 450 | 2.2960 | 0.1984 | 0.1655 | 0.1800 | 0.1984 |
72
+ | 4.1548 | 6.8493 | 500 | 2.1901 | 0.2381 | 0.2339 | 0.3055 | 0.2381 |
73
+ | 3.7031 | 7.5342 | 550 | 2.0601 | 0.3571 | 0.3299 | 0.4862 | 0.3571 |
74
+ | 3.4466 | 8.2192 | 600 | 2.0129 | 0.3651 | 0.3678 | 0.4649 | 0.3651 |
75
+ | 3.0481 | 8.9041 | 650 | 1.8144 | 0.4365 | 0.4292 | 0.4630 | 0.4365 |
76
+ | 2.5507 | 9.5890 | 700 | 1.6802 | 0.4921 | 0.4820 | 0.4894 | 0.4921 |
77
+ | 2.1803 | 10.2740 | 750 | 1.5281 | 0.5635 | 0.5703 | 0.6068 | 0.5635 |
78
+ | 1.7031 | 10.9589 | 800 | 1.4110 | 0.5714 | 0.5538 | 0.5561 | 0.5714 |
79
+ | 1.4117 | 11.6438 | 850 | 1.3102 | 0.6349 | 0.6388 | 0.6649 | 0.6349 |
80
+ | 1.0765 | 12.3288 | 900 | 1.2092 | 0.6746 | 0.6693 | 0.6834 | 0.6746 |
81
+ | 0.8571 | 13.0137 | 950 | 1.2143 | 0.6746 | 0.6679 | 0.6859 | 0.6746 |
82
+ | 0.6671 | 13.6986 | 1000 | 1.1043 | 0.6905 | 0.6811 | 0.6961 | 0.6905 |
83
+ | 0.5448 | 14.3836 | 1050 | 1.0635 | 0.7063 | 0.7057 | 0.7239 | 0.7063 |
84
+ | 0.419 | 15.0685 | 1100 | 1.0836 | 0.7381 | 0.7366 | 0.7522 | 0.7381 |
85
+ | 0.3435 | 15.7534 | 1150 | 1.0320 | 0.7063 | 0.7130 | 0.7487 | 0.7063 |
86
+ | 0.2654 | 16.4384 | 1200 | 1.0282 | 0.7063 | 0.7026 | 0.7209 | 0.7063 |
87
+ | 0.1986 | 17.1233 | 1250 | 1.0172 | 0.7063 | 0.7076 | 0.7218 | 0.7063 |
88
+ | 0.1714 | 17.8082 | 1300 | 1.0305 | 0.7302 | 0.7297 | 0.7572 | 0.7302 |
89
+ | 0.118 | 18.4932 | 1350 | 1.0045 | 0.7302 | 0.7293 | 0.7456 | 0.7302 |
90
+ | 0.1293 | 19.1781 | 1400 | 1.0415 | 0.7381 | 0.7402 | 0.7566 | 0.7381 |
91
+ | 0.0934 | 19.8630 | 1450 | 1.0429 | 0.7143 | 0.7183 | 0.7376 | 0.7143 |
92
+ | 0.0598 | 20.5479 | 1500 | 1.0438 | 0.7302 | 0.7310 | 0.7397 | 0.7302 |
93
+ | 0.0651 | 21.2329 | 1550 | 1.0299 | 0.7143 | 0.7187 | 0.7335 | 0.7143 |
94
+ | 0.0618 | 21.9178 | 1600 | 1.0538 | 0.7143 | 0.7185 | 0.7313 | 0.7143 |
95
+ | 0.0664 | 22.6027 | 1650 | 1.0280 | 0.7381 | 0.7394 | 0.7552 | 0.7381 |
96
+ | 0.0662 | 23.2877 | 1700 | 1.0319 | 0.7302 | 0.7320 | 0.7426 | 0.7302 |
97
+ | 0.0315 | 23.9726 | 1750 | 1.0467 | 0.7222 | 0.7251 | 0.7350 | 0.7222 |
98
+ | 0.0462 | 24.6575 | 1800 | 1.0465 | 0.7222 | 0.7251 | 0.7350 | 0.7222 |
99
 
100
 
101
  ### Framework versions
102
 
103
+ - Transformers 4.48.3
104
  - Pytorch 2.5.1+cu124
105
+ - Datasets 3.3.1
106
  - Tokenizers 0.21.0