Hugging Face's logo

arcleife
/
roberta-sentiment-id

Text Classification
Transformers
Safetensors
Indonesian
xlm-roberta
Model card Files Community
arcleife commited on
Commit
fbe9a85
·
verified ·
1 Parent(s): 205913f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -48,8 +48,8 @@ Refer [here](https://github.com/arcleife/notebooks/blob/main/sentiment_finetunin
48
  **Training hyperparameters**
49
 
50
  - num_train_epochs = 5
51
- - learning_rate = 1e-6
52
- - weight_decay = 5e-2
53
  - per_device_train_batch_size = 16
54
  - per_device_eval_batch_size = 16
55
  - fp16 = True
@@ -60,8 +60,8 @@ Refer [here](https://github.com/arcleife/notebooks/blob/main/sentiment_finetunin
60
 
61
  |Epoch| Training Loss | Validation Loss | F1 | Recall | Precision |
62
  |-----|---------------|-----------------|----------|----------|-----------|
63
- |1 | No log | 0.280201 | 0.926190 | 0.926190 | 0.926190 |
64
- |2 | No log | 0.251838 | 0.935714 | 0.935714 | 0.935714 |
65
- |3 | No log | 0.256298 | 0.940476 | 0.940476 | 0.940476 |
66
- |4 | No log | 0.304876 | 0.939683 | 0.939683 | 0.939683 |
67
- |5 | No log | 0.316441 | 0.941270 | 0.941270 | 0.941270 |
 
48
  **Training hyperparameters**
49
 
50
  - num_train_epochs = 5
51
+ - learning_rate = 5e-6
52
+ - weight_decay = 1e-1
53
  - per_device_train_batch_size = 16
54
  - per_device_eval_batch_size = 16
55
  - fp16 = True
 
60
 
61
  |Epoch| Training Loss | Validation Loss | F1 | Recall | Precision |
62
  |-----|---------------|-----------------|----------|----------|-----------|
63
+ |1 | No log | 0.283834 | 0.908730 | 0.908730 | 0.908730 |
64
+ |2 | No log | 0.248232 | 0.930952 | 0.930952 | 0.930952 |
65
+ |3 | No log | 0.282172 | 0.930952 | 0.930952 | 0.930952 |
66
+ |4 | No log | 0.257302 | 0.936508 | 0.936508 | 0.936508 |
67
+ |5 | No log | 0.271212 | 0.939683 | 0.939683 | 0.939683 |