Update README.md
Browse files
README.md
CHANGED
|
@@ -32,12 +32,13 @@ This model fine-tuned model of raygx/distilBERT-Nepali, revision no.: b35360e0cf
|
|
| 32 |
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
|
| 35 |
-
|
| 36 |
> - lowest: 17.31
|
| 37 |
> - average: 19.12z
|
| 38 |
|
| 39 |
(This is because training is done in batches of data due to limited resources available)
|
| 40 |
|
|
|
|
| 41 |
> - loss: 3.2503
|
| 42 |
> - val_loss: 3.0674
|
| 43 |
|
|
@@ -66,12 +67,13 @@ Perplexity:
|
|
| 66 |
- lowest: 17.31
|
| 67 |
- average: 19.12
|
| 68 |
|
| 69 |
-
|
| 70 |
-
loss:
|
| 71 |
-
loss: 3.
|
| 72 |
-
loss: 3.
|
| 73 |
-
loss: 3.
|
| 74 |
-
loss: 3.
|
|
|
|
| 75 |
|
| 76 |
### Framework versions
|
| 77 |
|
|
|
|
| 32 |
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
|
| 35 |
+
Perplexity:
|
| 36 |
> - lowest: 17.31
|
| 37 |
> - average: 19.12z
|
| 38 |
|
| 39 |
(This is because training is done in batches of data due to limited resources available)
|
| 40 |
|
| 41 |
+
Loss:
|
| 42 |
> - loss: 3.2503
|
| 43 |
> - val_loss: 3.0674
|
| 44 |
|
|
|
|
| 67 |
- lowest: 17.31
|
| 68 |
- average: 19.12
|
| 69 |
|
| 70 |
+
Loss:
|
| 71 |
+
- loss: 4.8605 - val_loss: 4.0510
|
| 72 |
+
- loss: 3.8504 - val_loss: 3.5142
|
| 73 |
+
- loss: 3.4918 - val_loss: 3.2408
|
| 74 |
+
- loss: 3.2503 - val_loss: 3.0674
|
| 75 |
+
- loss: 3.1324 - val_loss: 2.9243
|
| 76 |
+
- loss: 3.2503 - val_loss: 3.0674
|
| 77 |
|
| 78 |
### Framework versions
|
| 79 |
|