Hugging Face's logo

Intradiction
/
BERT-SA-LORA

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
Intradiction commited on
Commit
1417981
•
1 Parent(s): 961a566

End of training

Browse files
Files changed (2) hide show
  1. README.md +10 -15
  2. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.4183
22
- - Accuracy: 0.868
23
 
24
  ## Model description
25
 
@@ -40,32 +40,27 @@ More information needed
40
  The following hyperparameters were used during training:
41
  - learning_rate: 0.0002
42
  - train_batch_size: 8
43
- - eval_batch_size: 8
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
- - num_epochs: 10
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
52
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
53
- | No log | 1.0 | 125 | 0.6192 | 0.682 |
54
- | No log | 2.0 | 250 | 0.4579 | 0.79 |
55
- | No log | 3.0 | 375 | 0.3820 | 0.844 |
56
- | 0.4923 | 4.0 | 500 | 0.3872 | 0.865 |
57
- | 0.4923 | 5.0 | 625 | 0.3659 | 0.862 |
58
- | 0.4923 | 6.0 | 750 | 0.4615 | 0.853 |
59
- | 0.4923 | 7.0 | 875 | 0.4444 | 0.86 |
60
- | 0.2691 | 8.0 | 1000 | 0.4148 | 0.871 |
61
- | 0.2691 | 9.0 | 1125 | 0.4089 | 0.87 |
62
- | 0.2691 | 10.0 | 1250 | 0.4183 | 0.868 |
63
 
64
 
65
  ### Framework versions
66
 
67
  - PEFT 0.9.0
68
  - Transformers 4.38.2
69
- - Pytorch 2.2.1+cu121
70
  - Datasets 2.18.0
71
  - Tokenizers 0.15.2
 
18
 
19
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.3017
22
+ - Accuracy: 0.8943
23
 
24
  ## Model description
25
 
 
40
  The following hyperparameters were used during training:
41
  - learning_rate: 0.0002
42
  - train_batch_size: 8
43
+ - eval_batch_size: 16
44
  - seed: 42
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
+ - num_epochs: 5
48
 
49
  ### Training results
50
 
51
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
52
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
53
+ | No log | 1.0 | 375 | 0.3246 | 0.8687 |
54
+ | 0.5095 | 2.0 | 750 | 0.3154 | 0.8803 |
55
+ | 0.3046 | 3.0 | 1125 | 0.2957 | 0.888 |
56
+ | 0.2754 | 4.0 | 1500 | 0.2971 | 0.8937 |
57
+ | 0.2754 | 5.0 | 1875 | 0.3017 | 0.8943 |
 
 
 
 
 
58
 
59
 
60
  ### Framework versions
61
 
62
  - PEFT 0.9.0
63
  - Transformers 4.38.2
64
+ - Pytorch 2.2.1+cu118
65
  - Datasets 2.18.0
66
  - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b6e841db441569312ca3d1f9196406d4e9d00e323ef956d9bb9e26f136e56fd8
3
  size 1785816
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca525cf20248da3aaf5ddadd8e6dc070f9e017dd14635242e2a52ae9beca90df
3
  size 1785816