Update README.md
Browse files
README.md
CHANGED
|
@@ -5,4 +5,15 @@ license: mit
|
|
| 5 |
# DistilBERT
|
| 6 |
|
| 7 |
The DistilBERT model is a [BERT](https://huggingface.co/distilbert/distilbert-base-uncased) model fine-tuned on the
|
| 8 |
-
[NewsQA](https://huggingface.co/datasets/lucadiliello/newsqa) dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
# DistilBERT
|
| 6 |
|
| 7 |
The DistilBERT model is a [BERT](https://huggingface.co/distilbert/distilbert-base-uncased) model fine-tuned on the
|
| 8 |
+
[NewsQA](https://huggingface.co/datasets/lucadiliello/newsqa) dataset.
|
| 9 |
+
|
| 10 |
+
## Hyperparameters
|
| 11 |
+
```
|
| 12 |
+
batch_size = 16
|
| 13 |
+
n_epochs = 3
|
| 14 |
+
max_seq_len = 512
|
| 15 |
+
learning_rate = 2e-5
|
| 16 |
+
optimizer=Adam
|
| 17 |
+
weight_decay=0.01
|
| 18 |
+
embeds_dropout_prob = 0.1
|
| 19 |
+
```
|