Update README.md
Browse files
README.md
CHANGED
|
@@ -34,6 +34,17 @@ text = text.replace("ţ", "ț").replace("ş", "ș").replace("Ţ", "Ț").replace(
|
|
| 34 |
```
|
| 35 |
because the model was **NOT** trained on cedilla ``s`` and ``t``s. If you don't, you will have decreased performance due to ``<UNK>``s and increased number of tokens per word.
|
| 36 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
|
| 38 |
### Evaluation
|
| 39 |
|
|
|
|
| 34 |
```
|
| 35 |
because the model was **NOT** trained on cedilla ``s`` and ``t``s. If you don't, you will have decreased performance due to ``<UNK>``s and increased number of tokens per word.
|
| 36 |
|
| 37 |
+
### Parameters:
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
| Parameter | Value |
|
| 41 |
+
|------------------|-------|
|
| 42 |
+
| Batch size | 16 |
|
| 43 |
+
| Training steps | 256k |
|
| 44 |
+
| Warmup steps | 500 |
|
| 45 |
+
| Uncased | True |
|
| 46 |
+
| Max. Seq. Length | 512 |
|
| 47 |
+
|
| 48 |
|
| 49 |
### Evaluation
|
| 50 |
|