|
--- |
|
license: apache-2.0 |
|
base_model: google/flan-t5-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: t5-summarization-zero-shot-headers-and-better-prompt-base-enriched |
|
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. --> |
|
|
|
# t5-summarization-zero-shot-headers-and-better-prompt-base-enriched |
|
|
|
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: 3.3582 |
|
- Rouge: {'rouge1': 0.3973, 'rouge2': 0.1803, 'rougeL': 0.1995, 'rougeLsum': 0.1995} |
|
- Bert Score: 0.8772 |
|
- Bleurt 20: -0.7678 |
|
- Gen Len: 13.355 |
|
|
|
## 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: 0.0001 |
|
- 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 |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 20 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge | Bert Score | Bleurt 20 | Gen Len | |
|
|:-------------:|:-----:|:-----:|:---------------:|:---------------------------------------------------------------------------:|:----------:|:---------:|:-------:| |
|
| 2.188 | 1.0 | 601 | 2.1003 | {'rouge1': 0.4472, 'rouge2': 0.1969, 'rougeL': 0.1958, 'rougeLsum': 0.1958} | 0.8766 | -0.805 | 14.265 | |
|
| 1.8197 | 2.0 | 1202 | 1.9668 | {'rouge1': 0.4259, 'rouge2': 0.1977, 'rougeL': 0.2091, 'rougeLsum': 0.2091} | 0.8803 | -0.7854 | 13.395 | |
|
| 1.616 | 3.0 | 1803 | 1.9279 | {'rouge1': 0.4209, 'rouge2': 0.1984, 'rougeL': 0.2069, 'rougeLsum': 0.2069} | 0.8788 | -0.7915 | 13.385 | |
|
| 1.4174 | 4.0 | 2404 | 1.9601 | {'rouge1': 0.4294, 'rouge2': 0.2009, 'rougeL': 0.2098, 'rougeLsum': 0.2098} | 0.8796 | -0.7453 | 13.745 | |
|
| 1.2073 | 5.0 | 3005 | 1.9690 | {'rouge1': 0.3801, 'rouge2': 0.1813, 'rougeL': 0.2045, 'rougeLsum': 0.2045} | 0.8793 | -0.8024 | 12.63 | |
|
| 0.978 | 6.0 | 3606 | 2.1024 | {'rouge1': 0.4035, 'rouge2': 0.1887, 'rougeL': 0.2067, 'rougeLsum': 0.2067} | 0.8802 | -0.7427 | 13.08 | |
|
| 0.8994 | 7.0 | 4207 | 2.1300 | {'rouge1': 0.4209, 'rouge2': 0.1962, 'rougeL': 0.2063, 'rougeLsum': 0.2063} | 0.8821 | -0.7351 | 13.315 | |
|
| 0.8133 | 8.0 | 4808 | 2.2183 | {'rouge1': 0.4053, 'rouge2': 0.1857, 'rougeL': 0.2083, 'rougeLsum': 0.2083} | 0.8822 | -0.7597 | 13.105 | |
|
| 0.6993 | 9.0 | 5409 | 2.3794 | {'rouge1': 0.4158, 'rouge2': 0.1926, 'rougeL': 0.2056, 'rougeLsum': 0.2056} | 0.8789 | -0.762 | 13.73 | |
|
| 0.7033 | 10.0 | 6010 | 2.4450 | {'rouge1': 0.4119, 'rouge2': 0.1928, 'rougeL': 0.2059, 'rougeLsum': 0.2059} | 0.8804 | -0.7611 | 13.165 | |
|
| 0.5367 | 11.0 | 6611 | 2.6166 | {'rouge1': 0.3886, 'rouge2': 0.1776, 'rougeL': 0.1961, 'rougeLsum': 0.1961} | 0.8795 | -0.8055 | 12.925 | |
|
| 0.538 | 12.0 | 7212 | 2.6617 | {'rouge1': 0.3971, 'rouge2': 0.1762, 'rougeL': 0.1942, 'rougeLsum': 0.1942} | 0.878 | -0.7797 | 13.135 | |
|
| 0.5359 | 13.0 | 7813 | 2.8059 | {'rouge1': 0.4188, 'rouge2': 0.2008, 'rougeL': 0.209, 'rougeLsum': 0.209} | 0.8808 | -0.7481 | 13.445 | |
|
| 0.4019 | 14.0 | 8414 | 3.0293 | {'rouge1': 0.3901, 'rouge2': 0.1723, 'rougeL': 0.1972, 'rougeLsum': 0.1972} | 0.8765 | -0.7554 | 13.135 | |
|
| 0.3585 | 15.0 | 9015 | 3.0459 | {'rouge1': 0.405, 'rouge2': 0.1843, 'rougeL': 0.2023, 'rougeLsum': 0.2023} | 0.8789 | -0.7381 | 13.38 | |
|
| 0.3966 | 16.0 | 9616 | 3.0934 | {'rouge1': 0.392, 'rouge2': 0.176, 'rougeL': 0.1879, 'rougeLsum': 0.1879} | 0.8763 | -0.7684 | 13.18 | |
|
| 0.331 | 17.0 | 10217 | 3.1878 | {'rouge1': 0.406, 'rouge2': 0.1828, 'rougeL': 0.1975, 'rougeLsum': 0.1975} | 0.8771 | -0.7609 | 13.47 | |
|
| 0.3703 | 18.0 | 10818 | 3.2429 | {'rouge1': 0.4032, 'rouge2': 0.1798, 'rougeL': 0.197, 'rougeLsum': 0.197} | 0.8773 | -0.7613 | 13.465 | |
|
| 0.2751 | 19.0 | 11419 | 3.3337 | {'rouge1': 0.3983, 'rouge2': 0.1772, 'rougeL': 0.2009, 'rougeLsum': 0.2009} | 0.8778 | -0.7595 | 13.38 | |
|
| 0.2926 | 20.0 | 12020 | 3.3582 | {'rouge1': 0.3973, 'rouge2': 0.1803, 'rougeL': 0.1995, 'rougeLsum': 0.1995} | 0.8772 | -0.7678 | 13.355 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
|
|
