Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,23 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- Replete-AI/code_bagel
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
### Training hyperparameters
|
| 8 |
+
The following hyperparameters were used during training:
|
| 9 |
+
- learning_rate: 3e-05
|
| 10 |
+
- train_batch_size: 4
|
| 11 |
+
- eval_batch_size: 8
|
| 12 |
+
- seed: 42
|
| 13 |
+
- distributed_type: multi-GPU
|
| 14 |
+
- num_devices: 8
|
| 15 |
+
- gradient_accumulation_steps: 4
|
| 16 |
+
- total_train_batch_size: 128
|
| 17 |
+
- total_eval_batch_size: 64
|
| 18 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 19 |
+
- lr_scheduler_type: cosine
|
| 20 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 21 |
+
- num_epochs: 3.0
|
| 22 |
+
|
| 23 |
+
|