End of training
Browse files
README.md
CHANGED
|
@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# phi-1_5-medical_consultation
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on the
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
@@ -32,13 +32,17 @@ More information needed
|
|
| 32 |
### Training hyperparameters
|
| 33 |
|
| 34 |
The following hyperparameters were used during training:
|
| 35 |
-
- learning_rate:
|
| 36 |
- train_batch_size: 1
|
| 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 |
- num_epochs: 1
|
|
|
|
| 42 |
|
| 43 |
### Training results
|
| 44 |
|
|
|
|
| 13 |
|
| 14 |
# phi-1_5-medical_consultation
|
| 15 |
|
| 16 |
+
This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on the None dataset.
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
|
|
| 32 |
### Training hyperparameters
|
| 33 |
|
| 34 |
The following hyperparameters were used during training:
|
| 35 |
+
- learning_rate: 2e-05
|
| 36 |
- train_batch_size: 1
|
| 37 |
- eval_batch_size: 8
|
| 38 |
- seed: 42
|
| 39 |
+
- gradient_accumulation_steps: 2
|
| 40 |
+
- total_train_batch_size: 2
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: cosine
|
| 43 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 44 |
- num_epochs: 1
|
| 45 |
+
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|