Update README.md
Browse files
README.md
CHANGED
|
@@ -75,6 +75,7 @@ Whilst it would be beneficial to add the noise on the fly during training, to en
|
|
| 75 |
|
| 76 |
<details>
|
| 77 |
<summary> Hyperparameters </summary>
|
|
|
|
| 78 |
- learning_rate: 1e-05
|
| 79 |
- train_batch_size: 8
|
| 80 |
- eval_batch_size: 8
|
|
@@ -84,6 +85,7 @@ Whilst it would be beneficial to add the noise on the fly during training, to en
|
|
| 84 |
- lr_scheduler_warmup_steps: 500
|
| 85 |
- training_steps: 10000
|
| 86 |
- mixed_precision_training: Native AMP
|
|
|
|
| 87 |
</details>
|
| 88 |
|
| 89 |
### Training results
|
|
|
|
| 75 |
|
| 76 |
<details>
|
| 77 |
<summary> Hyperparameters </summary>
|
| 78 |
+
```txt
|
| 79 |
- learning_rate: 1e-05
|
| 80 |
- train_batch_size: 8
|
| 81 |
- eval_batch_size: 8
|
|
|
|
| 85 |
- lr_scheduler_warmup_steps: 500
|
| 86 |
- training_steps: 10000
|
| 87 |
- mixed_precision_training: Native AMP
|
| 88 |
+
```
|
| 89 |
</details>
|
| 90 |
|
| 91 |
### Training results
|