End of training
Browse files
README.md
CHANGED
|
@@ -13,7 +13,18 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# trained-distilbart-abs-0807
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [sshleifer/distilbart-xsum-12-6](https://huggingface.co/sshleifer/distilbart-xsum-12-6) on
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
## Model description
|
| 19 |
|
|
@@ -41,6 +52,18 @@ The following hyperparameters were used during training:
|
|
| 41 |
- num_epochs: 6
|
| 42 |
- mixed_precision_training: Native AMP
|
| 43 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
### Framework versions
|
| 45 |
|
| 46 |
- Transformers 4.41.2
|
|
|
|
| 13 |
|
| 14 |
# trained-distilbart-abs-0807
|
| 15 |
|
| 16 |
+
This model is a fine-tuned version of [sshleifer/distilbart-xsum-12-6](https://huggingface.co/sshleifer/distilbart-xsum-12-6) on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 2.4184
|
| 19 |
+
- Rouge/rouge1: 0.0185
|
| 20 |
+
- Rouge/rouge2: 0.0088
|
| 21 |
+
- Rouge/rougel: 0.0152
|
| 22 |
+
- Rouge/rougelsum: 0.016
|
| 23 |
+
- Bertscore/bertscore-precision: 0.0404
|
| 24 |
+
- Bertscore/bertscore-recall: 0.04
|
| 25 |
+
- Bertscore/bertscore-f1: 0.0402
|
| 26 |
+
- Meteor: 0.0163
|
| 27 |
+
- Gen Len: 80.0
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
|
|
| 52 |
- num_epochs: 6
|
| 53 |
- mixed_precision_training: Native AMP
|
| 54 |
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge/rouge1 | Rouge/rouge2 | Rouge/rougel | Rouge/rougelsum | Bertscore/bertscore-precision | Bertscore/bertscore-recall | Bertscore/bertscore-f1 | Meteor | Gen Len |
|
| 58 |
+
|:-------------:|:-----:|:----:|:---------------:|:------------:|:------------:|:------------:|:---------------:|:-----------------------------:|:--------------------------:|:----------------------:|:------:|:-------:|
|
| 59 |
+
| 2.105 | 1.0 | 220 | 2.0884 | 0.4545 | 0.2049 | 0.3881 | 0.3905 | 0.8969 | 0.88 | 0.8882 | 0.3923 | 80.0 |
|
| 60 |
+
| 1.8823 | 2.0 | 440 | 2.0066 | 0.3453 | 0.1575 | 0.2965 | 0.2984 | 0.6632 | 0.6547 | 0.6587 | 0.2949 | 80.0 |
|
| 61 |
+
| 1.4089 | 3.0 | 660 | 2.0717 | 0.0768 | 0.0337 | 0.0637 | 0.0639 | 0.1559 | 0.1535 | 0.1547 | 0.0667 | 80.0 |
|
| 62 |
+
| 1.0687 | 4.0 | 880 | 2.1627 | 0.0125 | 0.0048 | 0.0104 | 0.0114 | 0.0322 | 0.0317 | 0.0319 | 0.0118 | 80.0 |
|
| 63 |
+
| 0.7445 | 5.0 | 1100 | 2.2927 | 0.0402 | 0.0177 | 0.0332 | 0.0332 | 0.0815 | 0.0809 | 0.0812 | 0.0374 | 80.0 |
|
| 64 |
+
| 0.7619 | 6.0 | 1320 | 2.4184 | 0.0185 | 0.0088 | 0.0152 | 0.016 | 0.0404 | 0.04 | 0.0402 | 0.0163 | 80.0 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
### Framework versions
|
| 68 |
|
| 69 |
- Transformers 4.41.2
|