Update README.md
Browse files
README.md
CHANGED
|
@@ -22,8 +22,8 @@ This modelcard aims to be a base template for new models. It has been generated
|
|
| 22 |
|
| 23 |
|
| 24 |
|
| 25 |
-
- **Developed by:**
|
| 26 |
-
- **Finetuned from model
|
| 27 |
|
| 28 |
|
| 29 |
<!-- Provide the basic links for the model. -->
|
|
@@ -67,6 +67,11 @@ Using 2 datasets:
|
|
| 67 |
|
| 68 |
### Training Procedure
|
| 69 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
|
| 71 |
#### Training Hyperparameters
|
| 72 |
* Learning rate: 2e-5
|
|
@@ -80,6 +85,7 @@ Using 2 datasets:
|
|
| 80 |
| #2 | 0.9563 | 1.2139 | 0.74 |
|
| 81 |
| #3 | 0.3920 | 1.4264 | 0.75 |
|
| 82 |
| #4 | 0.2175 | 1.4742 | 0.74 |
|
|
|
|
| 83 |
I restored the check point in the 2nd epoch
|
| 84 |
|
| 85 |
|
|
|
|
| 22 |
|
| 23 |
|
| 24 |
|
| 25 |
+
- **Developed by:** Tô Hoàng Minh Tiến
|
| 26 |
+
- **Finetuned from model :** xml-roberta-base
|
| 27 |
|
| 28 |
|
| 29 |
<!-- Provide the basic links for the model. -->
|
|
|
|
| 67 |
|
| 68 |
### Training Procedure
|
| 69 |
|
| 70 |
+
#### Optimization:
|
| 71 |
+
* Adam
|
| 72 |
+
|
| 73 |
+
#### Loss function
|
| 74 |
+
* Cross entropy
|
| 75 |
|
| 76 |
#### Training Hyperparameters
|
| 77 |
* Learning rate: 2e-5
|
|
|
|
| 85 |
| #2 | 0.9563 | 1.2139 | 0.74 |
|
| 86 |
| #3 | 0.3920 | 1.4264 | 0.75 |
|
| 87 |
| #4 | 0.2175 | 1.4742 | 0.74 |
|
| 88 |
+
|
| 89 |
I restored the check point in the 2nd epoch
|
| 90 |
|
| 91 |
|