Commit ·
731f316
1
Parent(s): 1abd917
add to readme
Browse files
README.md
CHANGED
|
@@ -18,7 +18,7 @@ The *WordPiece* tokenizer uses several components:
|
|
| 18 |
|
| 19 |
## Training
|
| 20 |
|
| 21 |
-
Training was performed over 16M+ Dhivehi sentences/paragraphs. An Adam optimizer with weighted decay was used with following parameters:
|
| 22 |
|
| 23 |
* Learning rate: 1e-5
|
| 24 |
* Weight decay: 0.1
|
|
|
|
| 18 |
|
| 19 |
## Training
|
| 20 |
|
| 21 |
+
Training was performed over 16M+ Dhivehi sentences/paragraphs put together by [@ashraq](https://huggingface.co/ashraq). An Adam optimizer with weighted decay was used with following parameters:
|
| 22 |
|
| 23 |
* Learning rate: 1e-5
|
| 24 |
* Weight decay: 0.1
|