domenicrosati
commited on
Commit
•
482ab65
1
Parent(s):
f0f4a9c
update model card README.md
Browse files
README.md
CHANGED
@@ -16,11 +16,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
This model is a fine-tuned version of [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) on the None dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
-
- Loss:
|
20 |
-
- Rouge1: 0.
|
21 |
-
- Rouge2: 0.
|
22 |
-
- Rougel: 0.
|
23 |
-
- Rougelsum: 0.
|
24 |
|
25 |
## Model description
|
26 |
|
@@ -45,11 +45,17 @@ The following hyperparameters were used during training:
|
|
45 |
- seed: 42
|
46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
- lr_scheduler_type: linear
|
48 |
-
- num_epochs:
|
49 |
- mixed_precision_training: Native AMP
|
50 |
|
51 |
### Training results
|
52 |
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
|
54 |
|
55 |
### Framework versions
|
|
|
16 |
|
17 |
This model is a fine-tuned version of [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) on the None dataset.
|
18 |
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 2.3507
|
20 |
+
- Rouge1: 0.4610
|
21 |
+
- Rouge2: 0.1288
|
22 |
+
- Rougel: 0.2083
|
23 |
+
- Rougelsum: 0.2083
|
24 |
|
25 |
## Model description
|
26 |
|
|
|
45 |
- seed: 42
|
46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
- lr_scheduler_type: linear
|
48 |
+
- num_epochs: 4
|
49 |
- mixed_precision_training: Native AMP
|
50 |
|
51 |
### Training results
|
52 |
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
|
55 |
+
| 2.5385 | 0.92 | 1000 | 2.4379 | 0.4508 | 0.1237 | 0.2020 | 0.2021 |
|
56 |
+
| 2.2866 | 1.84 | 2000 | 2.3789 | 0.4512 | 0.1274 | 0.2063 | 0.2063 |
|
57 |
+
| 2.1157 | 2.76 | 3000 | 2.3597 | 0.4559 | 0.1268 | 0.2031 | 0.2032 |
|
58 |
+
| 1.9981 | 3.68 | 4000 | 2.3507 | 0.4610 | 0.1288 | 0.2083 | 0.2083 |
|
59 |
|
60 |
|
61 |
### Framework versions
|