Commit
•
3ff650d
1
Parent(s):
1d8e63d
Update README.md
Browse files
README.md
CHANGED
|
@@ -18,7 +18,7 @@ Warning: Since the model is trained on a large dataset, it may produce unethical
|
|
| 18 |
|
| 19 |
### Training Data
|
| 20 |
|
| 21 |
-
- Dataset size: ~
|
| 22 |
|
| 23 |
|
| 24 |
## Using model
|
|
@@ -65,9 +65,9 @@ print(generate_output("Türkiye'nin en çok tercih "))
|
|
| 65 |
|
| 66 |
#### Training Hyperparameters
|
| 67 |
|
| 68 |
-
- **Epochs:**
|
| 69 |
- **LearningRate:** 4e-4
|
| 70 |
|
| 71 |
|
| 72 |
#### Training Results
|
| 73 |
-
**training_loss:** 3.
|
|
|
|
| 18 |
|
| 19 |
### Training Data
|
| 20 |
|
| 21 |
+
- Dataset size: ~5 million data (Wikipedia, News and etc.)
|
| 22 |
|
| 23 |
|
| 24 |
## Using model
|
|
|
|
| 65 |
|
| 66 |
#### Training Hyperparameters
|
| 67 |
|
| 68 |
+
- **Epochs:** 10
|
| 69 |
- **LearningRate:** 4e-4
|
| 70 |
|
| 71 |
|
| 72 |
#### Training Results
|
| 73 |
+
**training_loss:** 3.5089332405925295
|