Update README.md
Browse files
README.md
CHANGED
|
@@ -47,9 +47,10 @@ Disclimer : The model output will be very based on the hyperparameter you're giv
|
|
| 47 |
### Model
|
| 48 |
* Architecture: a Transformer-based model with next-word prediction objective
|
| 49 |
* Dataset size: 50k samples
|
|
|
|
| 50 |
* Precision: fp16
|
| 51 |
* GPUs: 2xV100-16G
|
| 52 |
-
* Training time:
|
| 53 |
|
| 54 |
### Software
|
| 55 |
* [PyTorch](https://github.com/pytorch/pytorch)
|
|
|
|
| 47 |
### Model
|
| 48 |
* Architecture: a Transformer-based model with next-word prediction objective
|
| 49 |
* Dataset size: 50k samples
|
| 50 |
+
* Epochs: 5
|
| 51 |
* Precision: fp16
|
| 52 |
* GPUs: 2xV100-16G
|
| 53 |
+
* Training time: 2 days
|
| 54 |
|
| 55 |
### Software
|
| 56 |
* [PyTorch](https://github.com/pytorch/pytorch)
|