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
30b50a5
·
verified ·
1 Parent(s): 7b7d4da

Model save

Browse files
Files changed (3) hide show
  1. README.md +12 -21
  2. model.safetensors +1 -1
  3. runs/Apr22_12-30-43_tardis/events.out.tfevents.1713781942.tardis.82877.1 +3 -0
README.md CHANGED
@@ -3,9 +3,6 @@ license: apache-2.0
3
  base_model: google-bert/bert-base-cased
4
  tags:
5
  - generated_from_trainer
6
- metrics:
7
- - f1
8
- - recall
9
  model-index:
10
  - name: bert-base-cased
11
  results: []
@@ -18,14 +15,10 @@ 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: 0.3952
22
- - F1 Macro: 0.9092
23
- - F1: 0.9405
24
- - F1 Neg: 0.8779
25
- - Acc: 0.92
26
- - Prec: 0.9405
27
- - Recall: 0.9405
28
- - Mcc: 0.8184
29
 
30
  ## Model description
31
 
@@ -51,23 +44,21 @@ 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: 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
- | 0.2267 | 1.0 | 1647 | 0.4014 | 0.9184 | 0.9416 | 0.8951 | 0.925 | 0.9380 | 0.9453 | 0.8368 |
62
- | 0.0615 | 2.0 | 3294 | 0.6703 | 0.8792 | 0.9109 | 0.8475 | 0.8875 | 0.9237 | 0.8984 | 0.7590 |
63
- | 0.0252 | 3.0 | 4941 | 0.9033 | 0.8748 | 0.9105 | 0.8392 | 0.885 | 0.9070 | 0.9141 | 0.7497 |
64
- | 0.0121 | 4.0 | 6588 | 0.9735 | 0.8803 | 0.9144 | 0.8462 | 0.89 | 0.9109 | 0.9180 | 0.7606 |
65
- | 0.0018 | 5.0 | 8235 | 1.0559 | 0.8738 | 0.9069 | 0.8407 | 0.8825 | 0.9197 | 0.8945 | 0.7482 |
66
 
67
 
68
  ### Framework versions
69
 
70
- - Transformers 4.39.3
71
- - Pytorch 2.2.2+cu121
72
  - Datasets 2.18.0
73
  - Tokenizers 0.15.2
 
3
  base_model: google-bert/bert-base-cased
4
  tags:
5
  - generated_from_trainer
 
 
 
6
  model-index:
7
  - name: bert-base-cased
8
  results: []
 
15
 
16
  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.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.5549
19
+ - Icm: 0.2242
20
+ - Icmnorm: 0.6139
21
+ - Fmeasure: 0.7450
 
 
 
 
22
 
23
  ## Model description
24
 
 
44
  - distributed_type: multi-GPU
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
+ - num_epochs: 3
48
  - mixed_precision_training: Native AMP
49
 
50
  ### Training results
51
 
52
+ | Training Loss | Epoch | Step | Validation Loss | Icm | Icmnorm | Fmeasure |
53
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|
54
+ | No log | 1.0 | 193 | 0.5537 | 0.0351 | 0.5178 | 0.6717 |
55
+ | No log | 2.0 | 386 | 0.5549 | 0.2242 | 0.6139 | 0.7450 |
56
+ | 0.5445 | 3.0 | 579 | 0.6469 | 0.2201 | 0.6118 | 0.7433 |
 
 
57
 
58
 
59
  ### Framework versions
60
 
61
+ - Transformers 4.38.2
62
+ - Pytorch 2.2.1+cu121
63
  - Datasets 2.18.0
64
  - Tokenizers 0.15.2
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7fe8140df2f0a1822b772bb0b8dd6711a1dff94f7115bfc61746dc19002d0e3d
3
  size 433270768
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d726bc2411520a91f1a8bf0bb8223d0a04ff21bf380ed4959416f4125b46fa33
3
  size 433270768
runs/Apr22_12-30-43_tardis/events.out.tfevents.1713781942.tardis.82877.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f013895bdc0802c00b1aa279b118e7a94e319558793245a279cb3fd4728e9991
3
+ size 509