Hugging Face's logo

rendchevi
/
roberta-base-pr

Text Classification
Transformers
Safetensors
roberta
Generated from Trainer
text-embeddings-inference
Model card Files
xet
Community
rendchevi commited on
Commit
d91ae13
·
verified ·
1 Parent(s): 039239b

End of training

Browse files
Files changed (2) hide show
  1. README.md +16 -15
  2. model.safetensors +1 -1
README.md CHANGED
@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.8802
24
- - F1 Macro: 0.6182
25
- - Precision: 0.6222
26
- - Recall: 0.6202
27
- - Accuracy: 0.7742
28
 
29
  ## Model description
30
 
@@ -45,24 +45,25 @@ More information needed
45
  The following hyperparameters were used during training:
46
  - learning_rate: 2e-05
47
  - train_batch_size: 16
48
- - eval_batch_size: 16
49
  - seed: 42
50
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: linear
52
- - num_epochs: 8
 
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | F1 Macro | Precision | Recall | Accuracy |
57
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:|
58
- | No log | 1.0 | 240 | 0.8923 | 0.4815 | 0.5010 | 0.5028 | 0.7107 |
59
- | No log | 2.0 | 480 | 0.7876 | 0.5690 | 0.6087 | 0.5772 | 0.7477 |
60
- | 1.0781 | 3.0 | 720 | 0.7258 | 0.6166 | 0.6277 | 0.6229 | 0.7758 |
61
- | 1.0781 | 4.0 | 960 | 0.7538 | 0.6301 | 0.6368 | 0.6323 | 0.7804 |
62
- | 0.5204 | 5.0 | 1200 | 0.7760 | 0.6278 | 0.6424 | 0.6297 | 0.7825 |
63
- | 0.5204 | 6.0 | 1440 | 0.8245 | 0.6240 | 0.6264 | 0.6278 | 0.7789 |
64
- | 0.3145 | 7.0 | 1680 | 0.8588 | 0.6156 | 0.6189 | 0.6184 | 0.7726 |
65
- | 0.3145 | 8.0 | 1920 | 0.8802 | 0.6182 | 0.6222 | 0.6202 | 0.7742 |
66
 
67
 
68
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 1.1841
24
+ - F1 Macro: 0.6097
25
+ - Precision: 0.6135
26
+ - Recall: 0.6205
27
+ - Accuracy: 0.7627
28
 
29
  ## Model description
30
 
 
45
  The following hyperparameters were used during training:
46
  - learning_rate: 2e-05
47
  - train_batch_size: 16
48
+ - eval_batch_size: 32
49
  - seed: 42
50
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: linear
52
+ - lr_scheduler_warmup_ratio: 0.1
53
+ - num_epochs: 20
54
 
55
  ### Training results
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | F1 Macro | Precision | Recall | Accuracy |
58
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:|
59
+ | No log | 1.0 | 240 | 2.3617 | 0.0348 | 0.1449 | 0.1032 | 0.0749 |
60
+ | No log | 2.0 | 480 | 0.8375 | 0.5802 | 0.5865 | 0.6081 | 0.7399 |
61
+ | 1.9571 | 3.0 | 720 | 0.8221 | 0.5996 | 0.6040 | 0.6244 | 0.7471 |
62
+ | 1.9571 | 4.0 | 960 | 0.8073 | 0.6168 | 0.6096 | 0.6356 | 0.7617 |
63
+ | 0.9292 | 5.0 | 1200 | 0.7768 | 0.6273 | 0.6273 | 0.6369 | 0.7742 |
64
+ | 0.9292 | 6.0 | 1440 | 0.9650 | 0.6009 | 0.6025 | 0.6211 | 0.7445 |
65
+ | 0.5053 | 7.0 | 1680 | 1.0663 | 0.6072 | 0.6218 | 0.6186 | 0.7622 |
66
+ | 0.5053 | 8.0 | 1920 | 1.1841 | 0.6097 | 0.6135 | 0.6205 | 0.7627 |
67
 
68
 
69
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c924ce092b3a912249b9355feb9df3741f99ca4f7b2216d3face732071e7dd4f
3
  size 498640508
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f230e145aa7a98a137e0a8c0cf4e12ef87177897c5c5a9129489a92454bbbc6d
3
  size 498640508