Model save
Browse files
README.md
CHANGED
|
@@ -43,7 +43,7 @@ The following hyperparameters were used during training:
|
|
| 43 |
- gradient_accumulation_steps: 10
|
| 44 |
- total_train_batch_size: 10
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
-
- lr_scheduler_type:
|
| 47 |
- lr_scheduler_warmup_ratio: 0.03
|
| 48 |
- num_epochs: 3
|
| 49 |
|
|
|
|
| 43 |
- gradient_accumulation_steps: 10
|
| 44 |
- total_train_batch_size: 10
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: constant
|
| 47 |
- lr_scheduler_warmup_ratio: 0.03
|
| 48 |
- num_epochs: 3
|
| 49 |
|