|
--- |
|
base_model: google/flan-t5-base |
|
library_name: transformers |
|
license: apache-2.0 |
|
metrics: |
|
- rouge |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: flan-t5-rouge-durga-2 |
|
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-rouge-durga-2 |
|
|
|
This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0015 |
|
- Rouge1: 0.4073 |
|
- Rouge2: 0.3631 |
|
- Rougel: 0.4076 |
|
- Rougelsum: 0.4081 |
|
|
|
## 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: 0.0003 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- 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 | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:| |
|
| 3.6674 | 1.0 | 85 | 1.9307 | 0.2328 | 0.1210 | 0.2076 | 0.2079 | |
|
| 2.8171 | 2.0 | 170 | 1.5796 | 0.2548 | 0.1280 | 0.2250 | 0.2246 | |
|
| 2.1491 | 3.0 | 255 | 1.2905 | 0.2604 | 0.1445 | 0.2384 | 0.2376 | |
|
| 1.4061 | 4.0 | 340 | 1.0662 | 0.2631 | 0.1474 | 0.2447 | 0.2446 | |
|
| 1.0805 | 5.0 | 425 | 0.8724 | 0.2631 | 0.1479 | 0.2417 | 0.2417 | |
|
| 1.4852 | 6.0 | 510 | 0.7009 | 0.2861 | 0.1735 | 0.2718 | 0.2716 | |
|
| 1.7677 | 7.0 | 595 | 0.5283 | 0.2842 | 0.1876 | 0.2710 | 0.2708 | |
|
| 0.5014 | 8.0 | 680 | 0.3990 | 0.3142 | 0.2208 | 0.3011 | 0.3015 | |
|
| 1.5085 | 9.0 | 765 | 0.3122 | 0.3371 | 0.2522 | 0.3254 | 0.3255 | |
|
| 0.7696 | 10.0 | 850 | 0.2204 | 0.3537 | 0.2695 | 0.3423 | 0.3418 | |
|
| 0.7834 | 11.0 | 935 | 0.1808 | 0.3596 | 0.2813 | 0.3492 | 0.3491 | |
|
| 0.5238 | 12.0 | 1020 | 0.1300 | 0.3723 | 0.3043 | 0.3674 | 0.3674 | |
|
| 0.4623 | 13.0 | 1105 | 0.0940 | 0.3754 | 0.3166 | 0.3716 | 0.3716 | |
|
| 0.3645 | 14.0 | 1190 | 0.0679 | 0.3752 | 0.3149 | 0.3715 | 0.3727 | |
|
| 0.319 | 15.0 | 1275 | 0.0466 | 0.3902 | 0.3372 | 0.3888 | 0.3895 | |
|
| 0.1117 | 16.0 | 1360 | 0.0375 | 0.3945 | 0.3409 | 0.3928 | 0.3932 | |
|
| 0.1975 | 17.0 | 1445 | 0.0274 | 0.4010 | 0.3537 | 0.4002 | 0.4007 | |
|
| 0.2396 | 18.0 | 1530 | 0.0254 | 0.3985 | 0.3525 | 0.3976 | 0.3976 | |
|
| 0.3832 | 19.0 | 1615 | 0.0156 | 0.3976 | 0.3482 | 0.3963 | 0.3970 | |
|
| 0.2136 | 20.0 | 1700 | 0.0118 | 0.3998 | 0.3524 | 0.3996 | 0.4000 | |
|
| 0.1789 | 21.0 | 1785 | 0.0115 | 0.4022 | 0.3568 | 0.4020 | 0.4027 | |
|
| 0.0027 | 22.0 | 1870 | 0.0082 | 0.4047 | 0.3593 | 0.4051 | 0.4057 | |
|
| 0.2978 | 23.0 | 1955 | 0.0070 | 0.4028 | 0.3578 | 0.4032 | 0.4036 | |
|
| 0.1529 | 24.0 | 2040 | 0.0053 | 0.4075 | 0.3625 | 0.4078 | 0.4083 | |
|
| 0.0584 | 25.0 | 2125 | 0.0044 | 0.4073 | 0.3621 | 0.4075 | 0.4081 | |
|
| 0.0669 | 26.0 | 2210 | 0.0030 | 0.4071 | 0.3625 | 0.4073 | 0.4078 | |
|
| 0.2968 | 27.0 | 2295 | 0.0021 | 0.4073 | 0.3625 | 0.4074 | 0.4080 | |
|
| 0.072 | 28.0 | 2380 | 0.0018 | 0.4073 | 0.3625 | 0.4074 | 0.4080 | |
|
| 0.1676 | 29.0 | 2465 | 0.0016 | 0.4073 | 0.3631 | 0.4076 | 0.4081 | |
|
| 0.1567 | 30.0 | 2550 | 0.0015 | 0.4073 | 0.3631 | 0.4076 | 0.4081 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 3.0.0 |
|
- Tokenizers 0.19.1 |
|
|
