|
--- |
|
license: apache-2.0 |
|
tags: |
|
- summarization |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: t5-small-train |
|
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-small-train |
|
|
|
This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.2367 |
|
- Rouge1: 43.9525 |
|
- Rouge2: 22.3403 |
|
- Rougel: 38.7683 |
|
- Rougelsum: 39.2056 |
|
|
|
## 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: 4.6e-05 |
|
- train_batch_size: 9 |
|
- eval_batch_size: 9 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 8 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:| |
|
| 3.3237 | 1.0 | 40 | 2.6713 | 34.4731 | 14.9731 | 29.4814 | 29.9747 | |
|
| 2.7401 | 2.0 | 80 | 2.4318 | 38.1153 | 18.3492 | 33.4476 | 33.9181 | |
|
| 2.5882 | 3.0 | 120 | 2.3339 | 41.2707 | 19.8571 | 36.2685 | 36.6119 | |
|
| 2.4264 | 4.0 | 160 | 2.2878 | 42.184 | 20.9666 | 37.3488 | 37.6172 | |
|
| 2.3915 | 5.0 | 200 | 2.2605 | 43.4928 | 21.7195 | 38.4917 | 38.8471 | |
|
| 2.3599 | 6.0 | 240 | 2.2462 | 44.2876 | 22.28 | 38.9234 | 39.3673 | |
|
| 2.3073 | 7.0 | 280 | 2.2398 | 43.9822 | 22.3746 | 38.7625 | 39.0964 | |
|
| 2.3026 | 8.0 | 320 | 2.2367 | 43.9525 | 22.3403 | 38.7683 | 39.2056 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|
