Go Inoue commited on
Commit ·
ea2da6e
1
Parent(s): d953ee1
Fix typo
Browse files
README.md
CHANGED
|
@@ -120,7 +120,7 @@ We follow the original English BERT model's hyperparameters for pre-training, un
|
|
| 120 |
- We used Hugging Face's transformers to fine-tune our CAMeLBERT models.
|
| 121 |
- We used transformers `v3.1.0` along with PyTorch `v1.5.1`.
|
| 122 |
- The fine-tuning was done by adding a fully connected linear layer to the last hidden state.
|
| 123 |
-
- We use \\(F_{1}
|
| 124 |
- Code used for fine-tuning is available [here](https://github.com/CAMeL-Lab/CAMeLBERT).
|
| 125 |
|
| 126 |
### Results
|
|
|
|
| 120 |
- We used Hugging Face's transformers to fine-tune our CAMeLBERT models.
|
| 121 |
- We used transformers `v3.1.0` along with PyTorch `v1.5.1`.
|
| 122 |
- The fine-tuning was done by adding a fully connected linear layer to the last hidden state.
|
| 123 |
+
- We use \\(F_{1}\\) score as a metric for all tasks.
|
| 124 |
- Code used for fine-tuning is available [here](https://github.com/CAMeL-Lab/CAMeLBERT).
|
| 125 |
|
| 126 |
### Results
|