|
--- |
|
license: apache-2.0 |
|
base_model: google/flan-t5-small |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: fine-tune-llama2 |
|
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. --> |
|
|
|
# fine-tune-flan-t5 |
|
|
|
This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co./google/flan-t5-small) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.7668 |
|
- Rouge1: 0.267 |
|
- Rouge2: 0.1008 |
|
- Rougel: 0.2321 |
|
- Rougelsum: 0.2335 |
|
- Gen Len: 19.36 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- 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 | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:| |
|
| No log | 1.0 | 301 | 0.8071 | 0.2304 | 0.0799 | 0.1958 | 0.1966 | 19.42 | |
|
| 0.8375 | 2.0 | 602 | 0.8009 | 0.2368 | 0.0763 | 0.1966 | 0.1978 | 19.41 | |
|
| 0.8375 | 3.0 | 903 | 0.7935 | 0.2496 | 0.08 | 0.2114 | 0.2119 | 19.57 | |
|
| 0.8146 | 4.0 | 1204 | 0.7900 | 0.2556 | 0.0874 | 0.217 | 0.2178 | 19.55 | |
|
| 0.7923 | 5.0 | 1505 | 0.7893 | 0.2565 | 0.089 | 0.2145 | 0.2157 | 19.46 | |
|
| 0.7923 | 6.0 | 1806 | 0.7846 | 0.2512 | 0.087 | 0.2134 | 0.2139 | 19.35 | |
|
| 0.7801 | 7.0 | 2107 | 0.7845 | 0.2499 | 0.082 | 0.2082 | 0.209 | 19.34 | |
|
| 0.7801 | 8.0 | 2408 | 0.7792 | 0.2464 | 0.0822 | 0.2102 | 0.2105 | 19.52 | |
|
| 0.7589 | 9.0 | 2709 | 0.7769 | 0.2433 | 0.0774 | 0.2049 | 0.206 | 19.32 | |
|
| 0.7457 | 10.0 | 3010 | 0.7746 | 0.2459 | 0.0767 | 0.2073 | 0.2082 | 19.32 | |
|
| 0.7457 | 11.0 | 3311 | 0.7740 | 0.245 | 0.0817 | 0.2086 | 0.2087 | 19.32 | |
|
| 0.7495 | 12.0 | 3612 | 0.7736 | 0.2491 | 0.0824 | 0.2138 | 0.2144 | 19.31 | |
|
| 0.7495 | 13.0 | 3913 | 0.7716 | 0.2541 | 0.0876 | 0.2172 | 0.2182 | 19.29 | |
|
| 0.7341 | 14.0 | 4214 | 0.7694 | 0.2654 | 0.096 | 0.2274 | 0.2289 | 19.34 | |
|
| 0.711 | 15.0 | 4515 | 0.7677 | 0.2672 | 0.1054 | 0.23 | 0.2312 | 19.35 | |
|
| 0.711 | 16.0 | 4816 | 0.7698 | 0.2774 | 0.106 | 0.2395 | 0.2412 | 19.34 | |
|
| 0.7158 | 17.0 | 5117 | 0.7693 | 0.269 | 0.0973 | 0.2301 | 0.2317 | 19.5 | |
|
| 0.7158 | 18.0 | 5418 | 0.7683 | 0.2696 | 0.1062 | 0.233 | 0.2342 | 19.3 | |
|
| 0.7016 | 19.0 | 5719 | 0.7694 | 0.2601 | 0.0971 | 0.2258 | 0.2271 | 19.29 | |
|
| 0.6977 | 20.0 | 6020 | 0.7687 | 0.269 | 0.0999 | 0.2328 | 0.2341 | 19.33 | |
|
| 0.6977 | 21.0 | 6321 | 0.7676 | 0.2658 | 0.099 | 0.2297 | 0.2304 | 19.32 | |
|
| 0.7028 | 22.0 | 6622 | 0.7674 | 0.2654 | 0.0986 | 0.2299 | 0.2308 | 19.29 | |
|
| 0.7028 | 23.0 | 6923 | 0.7676 | 0.2666 | 0.0993 | 0.231 | 0.2328 | 19.34 | |
|
| 0.6698 | 24.0 | 7224 | 0.7667 | 0.2658 | 0.0989 | 0.2296 | 0.2314 | 19.35 | |
|
| 0.6956 | 25.0 | 7525 | 0.7670 | 0.2669 | 0.0992 | 0.2295 | 0.2308 | 19.35 | |
|
| 0.6956 | 26.0 | 7826 | 0.7669 | 0.2618 | 0.0966 | 0.2259 | 0.2269 | 19.31 | |
|
| 0.6776 | 27.0 | 8127 | 0.7671 | 0.2658 | 0.0983 | 0.2296 | 0.2319 | 19.35 | |
|
| 0.6776 | 28.0 | 8428 | 0.7672 | 0.2661 | 0.0998 | 0.2311 | 0.2328 | 19.35 | |
|
| 0.6891 | 29.0 | 8729 | 0.7668 | 0.267 | 0.1008 | 0.2321 | 0.2335 | 19.35 | |
|
| 0.6772 | 30.0 | 9030 | 0.7668 | 0.267 | 0.1008 | 0.2321 | 0.2335 | 19.36 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
