Hugging Face's logo

dtorber
/
bert-base-cased

Text Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Model card Files
xet
Metrics Community
dtorber commited on
Commit
a205f44
·
verified ·
1 Parent(s): 820c095

Model save

Browse files
Files changed (1) hide show
  1. README.md +14 -12
README.md CHANGED
@@ -18,14 +18,14 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [google-bert/bert-base-cased](https://huggingface.co/google-bert/bert-base-cased) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.4068
22
- - F1 Macro: 0.8879
23
- - F1: 0.9207
24
- - F1 Neg: 0.8551
25
- - Acc: 0.8975
26
- - Prec: 0.9119
27
- - Recall: 0.9297
28
- - Mcc: 0.7761
29
 
30
  ## Model description
31
 
@@ -51,16 +51,18 @@ The following hyperparameters were used during training:
51
  - distributed_type: multi-GPU
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: linear
54
- - num_epochs: 3
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 | F1 Neg | Acc | Prec | Recall | Mcc |
60
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:------:|
61
- | No log | 1.0 | 455 | 0.3953 | 0.8401 | 0.8681 | 0.8121 | 0.845 | 0.9533 | 0.7969 | 0.7001 |
62
- | 0.4578 | 2.0 | 910 | 0.3415 | 0.8924 | 0.9209 | 0.8639 | 0.9 | 0.932 | 0.9102 | 0.7854 |
63
- | 0.2713 | 3.0 | 1365 | 0.4068 | 0.8879 | 0.9207 | 0.8551 | 0.8975 | 0.9119 | 0.9297 | 0.7761 |
 
 
64
 
65
 
66
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [google-bert/bert-base-cased](https://huggingface.co/google-bert/bert-base-cased) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.5503
22
+ - F1 Macro: 0.9046
23
+ - F1: 0.9320
24
+ - F1 Neg: 0.8772
25
+ - Acc: 0.9125
26
+ - Prec: 0.9266
27
+ - Recall: 0.9375
28
+ - Mcc: 0.8094
29
 
30
  ## Model description
31
 
 
51
  - distributed_type: multi-GPU
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: linear
54
+ - num_epochs: 5
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
  | Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 | F1 Neg | Acc | Prec | Recall | Mcc |
60
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:------:|
61
+ | No log | 1.0 | 455 | 0.3613 | 0.8215 | 0.8706 | 0.7724 | 0.835 | 0.8740 | 0.8672 | 0.6430 |
62
+ | 0.4501 | 2.0 | 910 | 0.3824 | 0.8791 | 0.9193 | 0.8390 | 0.8925 | 0.8845 | 0.9570 | 0.7643 |
63
+ | 0.2737 | 3.0 | 1365 | 0.4971 | 0.8770 | 0.9084 | 0.8456 | 0.885 | 0.9268 | 0.8906 | 0.7552 |
64
+ | 0.1573 | 4.0 | 1820 | 0.5008 | 0.8912 | 0.9222 | 0.8601 | 0.9 | 0.9186 | 0.9258 | 0.7824 |
65
+ | 0.0839 | 5.0 | 2275 | 0.5503 | 0.9046 | 0.9320 | 0.8772 | 0.9125 | 0.9266 | 0.9375 | 0.8094 |
66
 
67
 
68
  ### Framework versions