|
--- |
|
license: apache-2.0 |
|
tags: |
|
- simplification |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: flan-t5-base-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. --> |
|
|
|
# flan-t5-base-clara-med |
|
|
|
This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.2699 |
|
- Rouge1: 30.1376 |
|
- Rouge2: 16.8424 |
|
- Rougel: 27.9649 |
|
- Rougelsum: 27.9946 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- 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 | 380 | 1.4710 | 27.6278 | 15.5057 | 25.9917 | 26.0601 | |
|
| No log | 2.0 | 760 | 1.3863 | 28.4324 | 15.8032 | 26.8023 | 26.8387 | |
|
| 1.6476 | 3.0 | 1140 | 1.3494 | 28.6807 | 16.0854 | 26.9253 | 26.9743 | |
|
| 1.6476 | 4.0 | 1520 | 1.3170 | 28.3434 | 15.6852 | 26.58 | 26.5937 | |
|
| 1.3695 | 5.0 | 1900 | 1.3009 | 28.8006 | 15.819 | 26.8122 | 26.8756 | |
|
| 1.3695 | 6.0 | 2280 | 1.2797 | 29.0521 | 16.4032 | 27.1802 | 27.1988 | |
|
| 1.3695 | 7.0 | 2660 | 1.2744 | 29.2339 | 16.4583 | 27.3799 | 27.4091 | |
|
| 1.2162 | 8.0 | 3040 | 1.2557 | 28.8177 | 16.2513 | 26.9967 | 27.028 | |
|
| 1.2162 | 9.0 | 3420 | 1.2553 | 29.0411 | 16.4606 | 27.2912 | 27.3004 | |
|
| 1.1232 | 10.0 | 3800 | 1.2540 | 29.0367 | 16.3896 | 27.2911 | 27.324 | |
|
| 1.1232 | 11.0 | 4180 | 1.2500 | 29.3928 | 16.6718 | 27.4638 | 27.4877 | |
|
| 1.1232 | 12.0 | 4560 | 1.2487 | 29.6046 | 16.7906 | 27.6814 | 27.6977 | |
|
| 1.0389 | 13.0 | 4940 | 1.2542 | 29.4922 | 16.5255 | 27.5363 | 27.5904 | |
|
| 1.0389 | 14.0 | 5320 | 1.2384 | 29.6472 | 16.707 | 27.6808 | 27.6988 | |
|
| 0.9794 | 15.0 | 5700 | 1.2476 | 29.3771 | 16.2381 | 27.3751 | 27.3876 | |
|
| 0.9794 | 16.0 | 6080 | 1.2437 | 29.4158 | 16.4003 | 27.3116 | 27.3409 | |
|
| 0.9794 | 17.0 | 6460 | 1.2466 | 29.2787 | 16.4136 | 27.3256 | 27.3622 | |
|
| 0.9276 | 18.0 | 6840 | 1.2530 | 29.4183 | 16.4244 | 27.325 | 27.3583 | |
|
| 0.9276 | 19.0 | 7220 | 1.2582 | 29.743 | 16.7631 | 27.6997 | 27.7752 | |
|
| 0.8851 | 20.0 | 7600 | 1.2560 | 29.5645 | 16.5834 | 27.5395 | 27.5622 | |
|
| 0.8851 | 21.0 | 7980 | 1.2544 | 29.4893 | 16.4478 | 27.3961 | 27.4465 | |
|
| 0.8851 | 22.0 | 8360 | 1.2593 | 29.785 | 16.6023 | 27.6214 | 27.6394 | |
|
| 0.8578 | 23.0 | 8740 | 1.2588 | 30.008 | 16.8796 | 27.882 | 27.8989 | |
|
| 0.8578 | 24.0 | 9120 | 1.2672 | 30.0112 | 16.6782 | 27.8556 | 27.8934 | |
|
| 0.8347 | 25.0 | 9500 | 1.2668 | 29.6945 | 16.431 | 27.4398 | 27.4956 | |
|
| 0.8347 | 26.0 | 9880 | 1.2642 | 29.9327 | 16.6105 | 27.798 | 27.8497 | |
|
| 0.8347 | 27.0 | 10260 | 1.2674 | 30.0747 | 16.7768 | 27.9137 | 27.9609 | |
|
| 0.8156 | 28.0 | 10640 | 1.2712 | 29.9504 | 16.6466 | 27.8371 | 27.8742 | |
|
| 0.8156 | 29.0 | 11020 | 1.2692 | 30.2209 | 16.9038 | 28.0454 | 28.0982 | |
|
| 0.8055 | 30.0 | 11400 | 1.2699 | 30.1376 | 16.8424 | 27.9649 | 27.9946 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.25.1 |
|
- Pytorch 1.13.0 |
|
- Datasets 2.8.0 |
|
- Tokenizers 0.12.1 |
|
|
