|
--- |
|
license: mit |
|
tags: |
|
- simplification |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: mbart-large-50-clara-med |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# mbart-large-50-clara-med |
|
|
|
This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.0952 |
|
- Rouge1: 49.4298 |
|
- Rouge2: 31.7193 |
|
- Rougel: 44.732 |
|
- Rougelsum: 44.9281 |
|
|
|
## Model description |
|
|
|
More information needed |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5.6e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:| |
|
| No log | 1.0 | 190 | 9.5151 | 8.9002 | 0.0056 | 8.9059 | 8.8991 | |
|
| No log | 2.0 | 380 | 1.7786 | 44.8765 | 27.9652 | 40.2081 | 40.3457 | |
|
| 4.488 | 3.0 | 570 | 1.7104 | 46.4054 | 28.8582 | 41.6579 | 41.86 | |
|
| 4.488 | 4.0 | 760 | 1.7601 | 47.6046 | 30.1854 | 42.9171 | 43.0745 | |
|
| 1.1057 | 5.0 | 950 | 1.9232 | 48.1693 | 30.1535 | 43.0418 | 43.1796 | |
|
| 1.1057 | 6.0 | 1140 | 2.2791 | 43.831 | 26.9216 | 39.1768 | 39.3672 | |
|
| 1.1057 | 7.0 | 1330 | 2.4800 | 42.4614 | 25.2371 | 37.6735 | 37.9309 | |
|
| 0.4401 | 8.0 | 1520 | 2.4991 | 46.6653 | 28.9836 | 42.1188 | 42.2492 | |
|
| 0.4401 | 9.0 | 1710 | 2.5826 | 47.2784 | 29.8703 | 42.622 | 42.7514 | |
|
| 0.2523 | 10.0 | 1900 | 2.6356 | 48.0382 | 30.8884 | 43.3523 | 43.5068 | |
|
| 0.2523 | 11.0 | 2090 | 2.6141 | 47.6911 | 29.3254 | 42.4938 | 42.6519 | |
|
| 0.2523 | 12.0 | 2280 | 2.6942 | 48.7597 | 30.9279 | 43.5391 | 43.6974 | |
|
| 0.1613 | 13.0 | 2470 | 2.7194 | 49.0916 | 30.9767 | 43.9943 | 44.1572 | |
|
| 0.1613 | 14.0 | 2660 | 2.7911 | 47.8223 | 30.6173 | 43.1809 | 43.3471 | |
|
| 0.1305 | 15.0 | 2850 | 2.8370 | 47.5629 | 29.7783 | 42.7168 | 42.8503 | |
|
| 0.1305 | 16.0 | 3040 | 2.8588 | 49.4762 | 31.6101 | 44.5422 | 44.7027 | |
|
| 0.1305 | 17.0 | 3230 | 2.9082 | 49.1502 | 31.4654 | 44.2166 | 44.3186 | |
|
| 0.141 | 18.0 | 3420 | 2.8887 | 48.9675 | 31.0485 | 44.177 | 44.3258 | |
|
| 0.141 | 19.0 | 3610 | 2.9043 | 49.2936 | 31.5204 | 44.2215 | 44.4216 | |
|
| 0.1096 | 20.0 | 3800 | 2.9549 | 48.0316 | 30.4505 | 42.9444 | 43.0893 | |
|
| 0.1096 | 21.0 | 3990 | 2.9666 | 49.2276 | 31.2755 | 44.2435 | 44.4591 | |
|
| 0.1096 | 22.0 | 4180 | 2.9697 | 49.1008 | 31.4931 | 44.1893 | 44.382 | |
|
| 0.0773 | 23.0 | 4370 | 2.9970 | 49.3707 | 31.4672 | 44.6066 | 44.7685 | |
|
| 0.0773 | 24.0 | 4560 | 3.0081 | 49.2172 | 31.4693 | 44.4235 | 44.5458 | |
|
| 0.048 | 25.0 | 4750 | 2.9968 | 49.4847 | 31.8341 | 44.8464 | 45.0286 | |
|
| 0.048 | 26.0 | 4940 | 3.0405 | 49.5724 | 31.612 | 44.5192 | 44.7717 | |
|
| 0.048 | 27.0 | 5130 | 3.0651 | 49.0194 | 31.2473 | 44.177 | 44.3837 | |
|
| 0.0274 | 28.0 | 5320 | 3.0740 | 49.2999 | 31.5672 | 44.56 | 44.8105 | |
|
| 0.0274 | 29.0 | 5510 | 3.0842 | 49.2898 | 31.602 | 44.5414 | 44.754 | |
|
| 0.0168 | 30.0 | 5700 | 3.0952 | 49.4298 | 31.7193 | 44.732 | 44.9281 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.25.1 |
|
- Pytorch 1.13.0 |
|
- Datasets 2.8.0 |
|
- Tokenizers 0.12.1 |
|
|
