Update README.md
Browse files
README.md
CHANGED
|
@@ -11,18 +11,6 @@ pipeline_tag: question-answering
|
|
| 11 |
|
| 12 |
This model is a fine-tuned version of [bert-base-chinese](https://huggingface.co/bert-base-chinese) on BD-11 dataset.
|
| 13 |
|
| 14 |
-
## Model description
|
| 15 |
-
|
| 16 |
-
More information needed
|
| 17 |
-
|
| 18 |
-
## Intended uses & limitations
|
| 19 |
-
|
| 20 |
-
More information needed
|
| 21 |
-
|
| 22 |
-
## Training and evaluation data
|
| 23 |
-
|
| 24 |
-
More information needed
|
| 25 |
-
|
| 26 |
## Training procedure
|
| 27 |
|
| 28 |
### Training hyperparameters
|
|
@@ -38,8 +26,18 @@ The following hyperparameters were used during training:
|
|
| 38 |
|
| 39 |
### Training results
|
| 40 |
|
| 41 |
-
|
| 42 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
|
| 44 |
### Framework versions
|
| 45 |
|
|
|
|
| 11 |
|
| 12 |
This model is a fine-tuned version of [bert-base-chinese](https://huggingface.co/bert-base-chinese) on BD-11 dataset.
|
| 13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
## Training procedure
|
| 15 |
|
| 16 |
### Training hyperparameters
|
|
|
|
| 26 |
|
| 27 |
### Training results
|
| 28 |
|
| 29 |
+
Evaluated on the BD-11 dataset test set.
|
| 30 |
+
```
|
| 31 |
+
"exact": 87.70,
|
| 32 |
+
"f1": 89.32,
|
| 33 |
+
"total": 1528,
|
| 34 |
+
"HasAns_exact": 83.94,
|
| 35 |
+
"HasAns_f1": 86.24,
|
| 36 |
+
"HasAns_total": 1077,
|
| 37 |
+
"NoAns_exact": 96.67,
|
| 38 |
+
"NoAns_f1": 96.67,
|
| 39 |
+
"NoAns_total": 451
|
| 40 |
+
```
|
| 41 |
|
| 42 |
### Framework versions
|
| 43 |
|