Update README.md
Browse files
README.md
CHANGED
|
@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
# BART-CNN-Convosumm
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on the arg-filtered reddit part of [Convosumm](https://github.com/Yale-LILY/ConvoSumm) dataset.
|
| 22 |
-
It achieves the following results on the evaluation set:
|
| 23 |
- Loss: 3.8797
|
| 24 |
- Rouge1: 38.6252
|
| 25 |
- Rouge2: 12.2556
|
|
@@ -27,6 +27,14 @@ It achieves the following results on the evaluation set:
|
|
| 27 |
- Rougelsum: 34.6324
|
| 28 |
- Gen Len: 81.28
|
| 29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
## Model description
|
| 31 |
|
| 32 |
More information needed
|
|
|
|
| 19 |
# BART-CNN-Convosumm
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on the arg-filtered reddit part of [Convosumm](https://github.com/Yale-LILY/ConvoSumm) dataset.
|
| 22 |
+
It achieves the following results on the evaluation set (50 data points):
|
| 23 |
- Loss: 3.8797
|
| 24 |
- Rouge1: 38.6252
|
| 25 |
- Rouge2: 12.2556
|
|
|
|
| 27 |
- Rougelsum: 34.6324
|
| 28 |
- Gen Len: 81.28
|
| 29 |
|
| 30 |
+
It achieves the following results on the test set (250 data points):
|
| 31 |
+
- Loss: 3.8343
|
| 32 |
+
- Rouge1: 38.3642
|
| 33 |
+
- Rouge2: 12.2056
|
| 34 |
+
- Rougel: 23.7782
|
| 35 |
+
- Rougelsum: 34.3959
|
| 36 |
+
- Gen Len: 84.132
|
| 37 |
+
|
| 38 |
## Model description
|
| 39 |
|
| 40 |
More information needed
|