Commit
·
471cc96
1
Parent(s):
a1a2ff1
End of training
Browse files
README.md
CHANGED
|
@@ -32,13 +32,13 @@ More information needed
|
|
| 32 |
### Training hyperparameters
|
| 33 |
|
| 34 |
The following hyperparameters were used during training:
|
| 35 |
-
- learning_rate:
|
| 36 |
-
- train_batch_size:
|
| 37 |
- eval_batch_size: 8
|
| 38 |
- seed: 42
|
| 39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
-
- lr_scheduler_type:
|
| 41 |
-
-
|
| 42 |
|
| 43 |
### Training results
|
| 44 |
|
|
@@ -48,5 +48,5 @@ The following hyperparameters were used during training:
|
|
| 48 |
|
| 49 |
- Transformers 4.34.1
|
| 50 |
- Pytorch 2.1.0+cu118
|
| 51 |
-
- Datasets 2.14.
|
| 52 |
- Tokenizers 0.14.1
|
|
|
|
| 32 |
### Training hyperparameters
|
| 33 |
|
| 34 |
The following hyperparameters were used during training:
|
| 35 |
+
- learning_rate: 0.0002
|
| 36 |
+
- train_batch_size: 4
|
| 37 |
- eval_batch_size: 8
|
| 38 |
- seed: 42
|
| 39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 40 |
+
- lr_scheduler_type: cosine
|
| 41 |
+
- training_steps: 1000
|
| 42 |
|
| 43 |
### Training results
|
| 44 |
|
|
|
|
| 48 |
|
| 49 |
- Transformers 4.34.1
|
| 50 |
- Pytorch 2.1.0+cu118
|
| 51 |
+
- Datasets 2.14.6
|
| 52 |
- Tokenizers 0.14.1
|