Hugging Face's logo

Heather-Driver
/
distilbert-classn-LinearAlg-finetuned-span-width-2

Transformers
TensorBoard
Safetensors
distilbert
Generated from Trainer
Model card Files Metrics Community
Heather-Driver commited on
Commit
a15561b
·
verified ·
1 Parent(s): acb6085

Classification Training

Browse files
Files changed (3) hide show
  1. README.md +52 -44
  2. model.safetensors +1 -1
  3. runs/Feb20_07-53-47_c8e62f6bf1d7/events.out.tfevents.1740038030.c8e62f6bf1d7.11498.0 +2 -2
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: 2.0778
25
- - Accuracy: 0.3333
26
- - F1: 0.3311
27
- - Precision: 0.3597
28
- - Recall: 0.3333
29
 
30
  ## Model description
31
 
@@ -44,55 +44,63 @@ More information needed
44
  ### Training hyperparameters
45
 
46
  The following hyperparameters were used during training:
47
- - learning_rate: 1.0000000000000008e-06
48
- - train_batch_size: 4
49
- - eval_batch_size: 4
50
  - seed: 42
 
 
51
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
  - lr_scheduler_type: linear
53
- - lr_scheduler_warmup_steps: 100
54
- - num_epochs: 30
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
60
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
61
- | 2.2243 | 1.0 | 73 | 2.3165 | 0.1984 | 0.1766 | 0.1716 | 0.1984 |
62
- | 2.168 | 2.0 | 146 | 2.3040 | 0.1984 | 0.1778 | 0.1719 | 0.1984 |
63
- | 2.2215 | 3.0 | 219 | 2.2916 | 0.2143 | 0.1919 | 0.1899 | 0.2143 |
64
- | 2.2083 | 4.0 | 292 | 2.2791 | 0.2302 | 0.2063 | 0.2055 | 0.2302 |
65
- | 2.1766 | 5.0 | 365 | 2.2713 | 0.2302 | 0.2126 | 0.2358 | 0.2302 |
66
- | 2.1182 | 6.0 | 438 | 2.2595 | 0.2222 | 0.2065 | 0.2164 | 0.2222 |
67
- | 2.1346 | 7.0 | 511 | 2.2410 | 0.2619 | 0.2424 | 0.2477 | 0.2619 |
68
- | 2.122 | 8.0 | 584 | 2.2372 | 0.2143 | 0.2020 | 0.2248 | 0.2143 |
69
- | 2.1259 | 9.0 | 657 | 2.2315 | 0.2381 | 0.2292 | 0.2400 | 0.2381 |
70
- | 2.1136 | 10.0 | 730 | 2.2180 | 0.2381 | 0.2275 | 0.2515 | 0.2381 |
71
- | 2.0655 | 11.0 | 803 | 2.2019 | 0.2778 | 0.2714 | 0.3003 | 0.2778 |
72
- | 2.0537 | 12.0 | 876 | 2.1831 | 0.2778 | 0.2687 | 0.2771 | 0.2778 |
73
- | 2.0474 | 13.0 | 949 | 2.1732 | 0.3016 | 0.2951 | 0.2976 | 0.3016 |
74
- | 2.0311 | 14.0 | 1022 | 2.1630 | 0.2937 | 0.2782 | 0.2688 | 0.2937 |
75
- | 2.0202 | 15.0 | 1095 | 2.1574 | 0.3016 | 0.2838 | 0.2733 | 0.3016 |
76
- | 2.0416 | 16.0 | 1168 | 2.1443 | 0.2937 | 0.2819 | 0.2751 | 0.2937 |
77
- | 1.9956 | 17.0 | 1241 | 2.1369 | 0.3254 | 0.3214 | 0.3543 | 0.3254 |
78
- | 1.9783 | 18.0 | 1314 | 2.1309 | 0.3095 | 0.3081 | 0.3223 | 0.3095 |
79
- | 1.9372 | 19.0 | 1387 | 2.1250 | 0.3254 | 0.3298 | 0.3500 | 0.3254 |
80
- | 1.9843 | 20.0 | 1460 | 2.1169 | 0.3413 | 0.3452 | 0.3653 | 0.3413 |
81
- | 1.9682 | 21.0 | 1533 | 2.1081 | 0.3492 | 0.3489 | 0.3873 | 0.3492 |
82
- | 1.9583 | 22.0 | 1606 | 2.0992 | 0.3730 | 0.3758 | 0.4085 | 0.3730 |
83
- | 1.9261 | 23.0 | 1679 | 2.0960 | 0.3413 | 0.3407 | 0.3708 | 0.3413 |
84
- | 1.8968 | 24.0 | 1752 | 2.0933 | 0.3413 | 0.3419 | 0.3714 | 0.3413 |
85
- | 1.8977 | 25.0 | 1825 | 2.0910 | 0.3571 | 0.3610 | 0.3887 | 0.3571 |
86
- | 1.9109 | 26.0 | 1898 | 2.0874 | 0.3413 | 0.3412 | 0.3705 | 0.3413 |
87
- | 1.893 | 27.0 | 1971 | 2.0800 | 0.3413 | 0.3382 | 0.3660 | 0.3413 |
88
- | 1.9244 | 28.0 | 2044 | 2.0792 | 0.3333 | 0.3311 | 0.3597 | 0.3333 |
89
- | 1.9181 | 29.0 | 2117 | 2.0779 | 0.3333 | 0.3311 | 0.3597 | 0.3333 |
90
- | 1.881 | 30.0 | 2190 | 2.0778 | 0.3333 | 0.3311 | 0.3597 | 0.3333 |
 
 
 
 
 
 
91
 
92
 
93
  ### Framework versions
94
 
95
- - Transformers 4.48.2
96
  - Pytorch 2.5.1+cu124
97
- - Datasets 3.2.0
98
  - 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: 1.0193
25
+ - Accuracy: 0.7778
26
+ - F1: 0.7763
27
+ - Precision: 0.8043
28
+ - Recall: 0.7778
29
 
30
  ## Model description
31
 
 
44
  ### Training hyperparameters
45
 
46
  The following hyperparameters were used during training:
47
+ - learning_rate: 1e-05
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
+ | 4.891 | 0.6849 | 50 | 2.4589 | 0.0952 | 0.0407 | 0.0293 | 0.0952 |
64
+ | 4.9428 | 1.3699 | 100 | 2.4348 | 0.1111 | 0.0722 | 0.0692 | 0.1111 |
65
+ | 4.8923 | 2.0548 | 150 | 2.4178 | 0.0873 | 0.0623 | 0.0532 | 0.0873 |
66
+ | 4.8117 | 2.7397 | 200 | 2.3912 | 0.1190 | 0.0977 | 0.0995 | 0.1190 |
67
+ | 4.7618 | 3.4247 | 250 | 2.4055 | 0.1190 | 0.0834 | 0.0792 | 0.1190 |
68
+ | 4.6817 | 4.1096 | 300 | 2.3647 | 0.1349 | 0.1036 | 0.1119 | 0.1349 |
69
+ | 4.6471 | 4.7945 | 350 | 2.3392 | 0.1190 | 0.0884 | 0.0977 | 0.1190 |
70
+ | 4.51 | 5.4795 | 400 | 2.2767 | 0.2063 | 0.1998 | 0.2209 | 0.2063 |
71
+ | 4.2849 | 6.1644 | 450 | 2.1959 | 0.2619 | 0.2654 | 0.3733 | 0.2619 |
72
+ | 4.0971 | 6.8493 | 500 | 2.1070 | 0.3413 | 0.3294 | 0.3556 | 0.3413 |
73
+ | 3.5719 | 7.5342 | 550 | 2.0605 | 0.3016 | 0.2690 | 0.2709 | 0.3016 |
74
+ | 3.4648 | 8.2192 | 600 | 1.8455 | 0.4603 | 0.4470 | 0.4697 | 0.4603 |
75
+ | 3.0697 | 8.9041 | 650 | 1.7117 | 0.5238 | 0.5097 | 0.5376 | 0.5238 |
76
+ | 2.472 | 9.5890 | 700 | 1.5777 | 0.5159 | 0.5045 | 0.5396 | 0.5159 |
77
+ | 2.1915 | 10.2740 | 750 | 1.4357 | 0.5317 | 0.5294 | 0.5559 | 0.5317 |
78
+ | 1.7396 | 10.9589 | 800 | 1.2750 | 0.6667 | 0.6612 | 0.6873 | 0.6667 |
79
+ | 1.4216 | 11.6438 | 850 | 1.1920 | 0.6667 | 0.6685 | 0.6815 | 0.6667 |
80
+ | 1.1105 | 12.3288 | 900 | 1.1391 | 0.6825 | 0.6779 | 0.7026 | 0.6825 |
81
+ | 0.9306 | 13.0137 | 950 | 1.1158 | 0.6667 | 0.6632 | 0.6854 | 0.6667 |
82
+ | 0.7232 | 13.6986 | 1000 | 1.0369 | 0.7460 | 0.7475 | 0.7636 | 0.7460 |
83
+ | 0.5534 | 14.3836 | 1050 | 0.9872 | 0.7381 | 0.7365 | 0.7528 | 0.7381 |
84
+ | 0.4906 | 15.0685 | 1100 | 1.0500 | 0.7381 | 0.7382 | 0.7523 | 0.7381 |
85
+ | 0.3192 | 15.7534 | 1150 | 1.0005 | 0.7381 | 0.7398 | 0.7499 | 0.7381 |
86
+ | 0.3181 | 16.4384 | 1200 | 0.9607 | 0.7698 | 0.7682 | 0.7753 | 0.7698 |
87
+ | 0.2476 | 17.1233 | 1250 | 0.9588 | 0.7619 | 0.7637 | 0.7813 | 0.7619 |
88
+ | 0.2185 | 17.8082 | 1300 | 1.0287 | 0.7460 | 0.7395 | 0.7641 | 0.7460 |
89
+ | 0.1517 | 18.4932 | 1350 | 0.9674 | 0.7698 | 0.7683 | 0.7864 | 0.7698 |
90
+ | 0.1507 | 19.1781 | 1400 | 0.9744 | 0.7698 | 0.7662 | 0.7796 | 0.7698 |
91
+ | 0.1135 | 19.8630 | 1450 | 1.0068 | 0.7619 | 0.7598 | 0.7806 | 0.7619 |
92
+ | 0.0871 | 20.5479 | 1500 | 1.0009 | 0.7698 | 0.7688 | 0.7872 | 0.7698 |
93
+ | 0.0676 | 21.2329 | 1550 | 0.9972 | 0.7778 | 0.7743 | 0.7883 | 0.7778 |
94
+ | 0.0916 | 21.9178 | 1600 | 1.0069 | 0.7778 | 0.7748 | 0.7929 | 0.7778 |
95
+ | 0.0848 | 22.6027 | 1650 | 1.0176 | 0.7778 | 0.7750 | 0.8011 | 0.7778 |
96
+ | 0.0739 | 23.2877 | 1700 | 1.0196 | 0.7698 | 0.7686 | 0.7968 | 0.7698 |
97
+ | 0.0455 | 23.9726 | 1750 | 1.0198 | 0.7698 | 0.7693 | 0.7976 | 0.7698 |
98
+ | 0.0572 | 24.6575 | 1800 | 1.0193 | 0.7778 | 0.7763 | 0.8043 | 0.7778 |
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
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:94e244b9551638e56a1e3db3e38dd9924e293c7117eae42318f4206b0edbc6db
3
  size 262453592
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77c729fa696dc3e6748adcb3cf40aa6dc91db3cb19a1561c965640aae7b9cde4
3
  size 262453592
runs/Feb20_07-53-47_c8e62f6bf1d7/events.out.tfevents.1740038030.c8e62f6bf1d7.11498.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ad114468e6031967b41c3f13b2a07bedb5f4e652d73e8b1a6704464974739fa8
3
- size 26796
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8db95d0c287e3a92efedd8da0133fd50483db80de0d2d4612e81e470854d0ea
3
+ size 30565