update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- simplification
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- rouge
|
| 7 |
+
model-index:
|
| 8 |
+
- name: pegasus-xsum-clara-med
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# pegasus-xsum-clara-med
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [google/pegasus-xsum](https://huggingface.co/google/pegasus-xsum) on the None dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.9013
|
| 20 |
+
- Rouge1: 43.7745
|
| 21 |
+
- Rouge2: 25.6611
|
| 22 |
+
- Rougel: 39.6651
|
| 23 |
+
- Rougelsum: 39.7513
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 5.6e-05
|
| 43 |
+
- train_batch_size: 16
|
| 44 |
+
- eval_batch_size: 16
|
| 45 |
+
- seed: 42
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: linear
|
| 48 |
+
- num_epochs: 30
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
| 53 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
|
| 54 |
+
| No log | 1.0 | 190 | 2.5468 | 41.5429 | 24.0669 | 37.7315 | 37.8487 |
|
| 55 |
+
| No log | 2.0 | 380 | 2.3603 | 41.9098 | 24.2645 | 38.0814 | 38.1943 |
|
| 56 |
+
| 2.7787 | 3.0 | 570 | 2.2604 | 41.9695 | 24.4377 | 38.1431 | 38.2448 |
|
| 57 |
+
| 2.7787 | 4.0 | 760 | 2.1846 | 42.0768 | 24.5895 | 38.3383 | 38.467 |
|
| 58 |
+
| 2.2691 | 5.0 | 950 | 2.1361 | 42.3917 | 24.843 | 38.5836 | 38.7019 |
|
| 59 |
+
| 2.2691 | 6.0 | 1140 | 2.0887 | 42.5399 | 24.8959 | 38.662 | 38.8238 |
|
| 60 |
+
| 2.2691 | 7.0 | 1330 | 2.0617 | 42.7411 | 24.9073 | 38.9019 | 39.0312 |
|
| 61 |
+
| 2.0313 | 8.0 | 1520 | 2.0222 | 43.0061 | 25.3295 | 39.13 | 39.2656 |
|
| 62 |
+
| 2.0313 | 9.0 | 1710 | 2.0049 | 43.225 | 25.4193 | 39.3897 | 39.4901 |
|
| 63 |
+
| 1.872 | 10.0 | 1900 | 1.9899 | 43.2609 | 25.4765 | 39.3409 | 39.4977 |
|
| 64 |
+
| 1.872 | 11.0 | 2090 | 1.9772 | 43.4282 | 25.7581 | 39.5309 | 39.6755 |
|
| 65 |
+
| 1.872 | 12.0 | 2280 | 1.9630 | 43.5876 | 25.6747 | 39.5548 | 39.6953 |
|
| 66 |
+
| 1.7497 | 13.0 | 2470 | 1.9513 | 43.392 | 25.5144 | 39.4745 | 39.5719 |
|
| 67 |
+
| 1.7497 | 14.0 | 2660 | 1.9336 | 43.2143 | 25.4173 | 39.3053 | 39.3998 |
|
| 68 |
+
| 1.6609 | 15.0 | 2850 | 1.9345 | 43.2735 | 25.5504 | 39.3319 | 39.4607 |
|
| 69 |
+
| 1.6609 | 16.0 | 3040 | 1.9152 | 43.4389 | 25.5619 | 39.4289 | 39.5489 |
|
| 70 |
+
| 1.6609 | 17.0 | 3230 | 1.9106 | 43.2504 | 25.2506 | 39.2615 | 39.3613 |
|
| 71 |
+
| 1.5809 | 18.0 | 3420 | 1.9125 | 43.2177 | 25.2772 | 39.2781 | 39.3551 |
|
| 72 |
+
| 1.5809 | 19.0 | 3610 | 1.9086 | 43.1484 | 25.262 | 39.2296 | 39.2846 |
|
| 73 |
+
| 1.5221 | 20.0 | 3800 | 1.9030 | 43.299 | 25.3763 | 39.3083 | 39.3957 |
|
| 74 |
+
| 1.5221 | 21.0 | 3990 | 1.8996 | 43.1728 | 25.3052 | 39.2469 | 39.3093 |
|
| 75 |
+
| 1.5221 | 22.0 | 4180 | 1.9006 | 43.4838 | 25.3932 | 39.3524 | 39.4185 |
|
| 76 |
+
| 1.4714 | 23.0 | 4370 | 1.8977 | 43.5923 | 25.554 | 39.4536 | 39.5537 |
|
| 77 |
+
| 1.4714 | 24.0 | 4560 | 1.9035 | 43.6778 | 25.6253 | 39.5759 | 39.6767 |
|
| 78 |
+
| 1.4421 | 25.0 | 4750 | 1.9000 | 43.4975 | 25.5578 | 39.4462 | 39.5438 |
|
| 79 |
+
| 1.4421 | 26.0 | 4940 | 1.9030 | 43.4647 | 25.5185 | 39.4997 | 39.5935 |
|
| 80 |
+
| 1.4421 | 27.0 | 5130 | 1.8993 | 43.3463 | 25.4989 | 39.4247 | 39.5183 |
|
| 81 |
+
| 1.4139 | 28.0 | 5320 | 1.9009 | 43.5812 | 25.6923 | 39.6128 | 39.6974 |
|
| 82 |
+
| 1.4139 | 29.0 | 5510 | 1.9002 | 43.724 | 25.655 | 39.7022 | 39.8188 |
|
| 83 |
+
| 1.4016 | 30.0 | 5700 | 1.9013 | 43.7745 | 25.6611 | 39.6651 | 39.7513 |
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
### Framework versions
|
| 87 |
+
|
| 88 |
+
- Transformers 4.25.1
|
| 89 |
+
- Pytorch 1.13.0
|
| 90 |
+
- Datasets 2.8.0
|
| 91 |
+
- Tokenizers 0.12.1
|