|
--- |
|
license: apache-2.0 |
|
base_model: google/flan-t5-small |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: t5-summarization-one-shot-better-prompt |
|
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-one-shot-better-prompt |
|
|
|
This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co/google/flan-t5-small) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.2431 |
|
- Rouge: {'rouge1': 39.1164, 'rouge2': 19.0784, 'rougeL': 20.2856, 'rougeLsum': 20.2856} |
|
- Bert Score: 0.8802 |
|
- Bleurt 20: -0.7688 |
|
- Gen Len: 13.545 |
|
|
|
## 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: 7 |
|
- eval_batch_size: 7 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 20 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge | Bert Score | Bleurt 20 | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------------------------------------------------------------------------------:|:----------:|:---------:|:-------:| |
|
| 2.6921 | 1.0 | 172 | 2.4379 | {'rouge1': 43.7984, 'rouge2': 17.2952, 'rougeL': 18.5604, 'rougeLsum': 18.5604} | 0.869 | -0.8739 | 14.84 | |
|
| 2.5119 | 2.0 | 344 | 2.3282 | {'rouge1': 41.5219, 'rouge2': 17.4612, 'rougeL': 19.5103, 'rougeLsum': 19.5103} | 0.8749 | -0.8329 | 13.7 | |
|
| 2.3033 | 3.0 | 516 | 2.2821 | {'rouge1': 41.0636, 'rouge2': 18.2347, 'rougeL': 19.8704, 'rougeLsum': 19.8704} | 0.878 | -0.8268 | 13.75 | |
|
| 2.2139 | 4.0 | 688 | 2.2404 | {'rouge1': 39.9679, 'rouge2': 18.8795, 'rougeL': 19.7032, 'rougeLsum': 19.7032} | 0.8796 | -0.8035 | 13.305 | |
|
| 2.0835 | 5.0 | 860 | 2.2446 | {'rouge1': 41.8958, 'rouge2': 18.439, 'rougeL': 19.2982, 'rougeLsum': 19.2982} | 0.877 | -0.7963 | 14.34 | |
|
| 2.0379 | 6.0 | 1032 | 2.2233 | {'rouge1': 40.9703, 'rouge2': 19.7574, 'rougeL': 19.9387, 'rougeLsum': 19.9387} | 0.8793 | -0.7805 | 13.625 | |
|
| 1.959 | 7.0 | 1204 | 2.2073 | {'rouge1': 39.2194, 'rouge2': 18.9553, 'rougeL': 19.7847, 'rougeLsum': 19.7847} | 0.8787 | -0.8045 | 13.365 | |
|
| 1.9177 | 8.0 | 1376 | 2.2146 | {'rouge1': 40.8391, 'rouge2': 19.5219, 'rougeL': 20.2602, 'rougeLsum': 20.2602} | 0.8781 | -0.7974 | 13.865 | |
|
| 1.8749 | 9.0 | 1548 | 2.2071 | {'rouge1': 40.9497, 'rouge2': 19.9867, 'rougeL': 20.5682, 'rougeLsum': 20.5682} | 0.8808 | -0.7812 | 13.68 | |
|
| 1.8112 | 10.0 | 1720 | 2.2045 | {'rouge1': 36.465, 'rouge2': 16.4287, 'rougeL': 19.1978, 'rougeLsum': 19.1978} | 0.8772 | -0.8384 | 13.295 | |
|
| 1.7475 | 11.0 | 1892 | 2.2210 | {'rouge1': 39.4889, 'rouge2': 19.1309, 'rougeL': 19.879, 'rougeLsum': 19.879} | 0.8785 | -0.8074 | 13.585 | |
|
| 1.7384 | 12.0 | 2064 | 2.2269 | {'rouge1': 38.2904, 'rouge2': 18.2873, 'rougeL': 19.4418, 'rougeLsum': 19.4418} | 0.8789 | -0.7984 | 13.42 | |
|
| 1.6849 | 13.0 | 2236 | 2.2261 | {'rouge1': 37.6283, 'rouge2': 17.6979, 'rougeL': 19.584, 'rougeLsum': 19.584} | 0.878 | -0.7885 | 13.445 | |
|
| 1.6531 | 14.0 | 2408 | 2.2186 | {'rouge1': 38.7975, 'rouge2': 19.0939, 'rougeL': 20.7873, 'rougeLsum': 20.7873} | 0.8806 | -0.783 | 13.445 | |
|
| 1.663 | 15.0 | 2580 | 2.2245 | {'rouge1': 38.9159, 'rouge2': 19.153, 'rougeL': 20.5232, 'rougeLsum': 20.5232} | 0.8811 | -0.7514 | 13.59 | |
|
| 1.6036 | 16.0 | 2752 | 2.2430 | {'rouge1': 37.6184, 'rouge2': 17.6773, 'rougeL': 19.2693, 'rougeLsum': 19.2693} | 0.8771 | -0.7992 | 13.6 | |
|
| 1.6333 | 17.0 | 2924 | 2.2418 | {'rouge1': 38.1301, 'rouge2': 18.4061, 'rougeL': 20.1355, 'rougeLsum': 20.1355} | 0.879 | -0.7845 | 13.49 | |
|
| 1.6322 | 18.0 | 3096 | 2.2421 | {'rouge1': 38.0746, 'rouge2': 18.2039, 'rougeL': 19.7404, 'rougeLsum': 19.7404} | 0.8789 | -0.7892 | 13.41 | |
|
| 1.5982 | 19.0 | 3268 | 2.2411 | {'rouge1': 39.1375, 'rouge2': 19.1696, 'rougeL': 20.2695, 'rougeLsum': 20.2695} | 0.8802 | -0.7713 | 13.465 | |
|
| 1.593 | 20.0 | 3440 | 2.2431 | {'rouge1': 39.1164, 'rouge2': 19.0784, 'rougeL': 20.2856, 'rougeLsum': 20.2856} | 0.8802 | -0.7688 | 13.545 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
|
|
