|
--- |
|
license: apache-2.0 |
|
tags: |
|
- summarization |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: mt5-small-text-sum-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. --> |
|
|
|
# mt5-small-text-sum-2 |
|
|
|
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.3612 |
|
- Rouge1: 21.38 |
|
- Rouge2: 6.57 |
|
- Rougel: 21.08 |
|
|
|
## 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: 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: 40 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | |
|
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:| |
|
| 4.7204 | 1.45 | 500 | 2.6053 | 16.9 | 4.9 | 16.73 | |
|
| 3.1289 | 2.9 | 1000 | 2.4878 | 17.96 | 5.26 | 17.82 | |
|
| 2.8862 | 4.35 | 1500 | 2.4109 | 17.4 | 5.08 | 17.14 | |
|
| 2.7669 | 5.8 | 2000 | 2.4006 | 18.53 | 5.29 | 18.21 | |
|
| 2.6433 | 7.25 | 2500 | 2.4017 | 18.69 | 5.71 | 18.53 | |
|
| 2.5514 | 8.7 | 3000 | 2.3917 | 19.32 | 5.89 | 19.12 | |
|
| 2.4947 | 10.14 | 3500 | 2.3994 | 20.56 | 6.08 | 20.19 | |
|
| 2.3995 | 11.59 | 4000 | 2.3608 | 20.11 | 6.52 | 19.75 | |
|
| 2.3798 | 13.04 | 4500 | 2.3251 | 19.98 | 6.26 | 19.76 | |
|
| 2.3029 | 14.49 | 5000 | 2.3387 | 19.71 | 6.11 | 19.42 | |
|
| 2.2563 | 15.94 | 5500 | 2.3372 | 20.18 | 6.34 | 19.8 | |
|
| 2.2109 | 17.39 | 6000 | 2.3410 | 20.58 | 6.35 | 20.14 | |
|
| 2.166 | 18.84 | 6500 | 2.3432 | 20.93 | 6.5 | 20.63 | |
|
| 2.1283 | 20.29 | 7000 | 2.3404 | 21.0 | 6.5 | 20.73 | |
|
| 2.1054 | 21.74 | 7500 | 2.3563 | 20.95 | 6.54 | 20.48 | |
|
| 2.0658 | 23.19 | 8000 | 2.3575 | 19.73 | 6.18 | 19.54 | |
|
| 2.0461 | 24.64 | 8500 | 2.3382 | 20.78 | 6.42 | 20.52 | |
|
| 2.0135 | 26.09 | 9000 | 2.3628 | 20.94 | 6.55 | 20.66 | |
|
| 2.0122 | 27.54 | 9500 | 2.3725 | 21.1 | 6.87 | 20.96 | |
|
| 1.9623 | 28.99 | 10000 | 2.3612 | 21.38 | 6.57 | 21.08 | |
|
| 1.9518 | 30.43 | 10500 | 2.3619 | 20.12 | 6.25 | 19.8 | |
|
| 1.9327 | 31.88 | 11000 | 2.3642 | 20.9 | 6.6 | 20.55 | |
|
| 1.9147 | 33.33 | 11500 | 2.3703 | 21.0 | 6.37 | 20.59 | |
|
| 1.9145 | 34.78 | 12000 | 2.3823 | 21.24 | 6.84 | 20.92 | |
|
| 1.9065 | 36.23 | 12500 | 2.3686 | 20.16 | 6.41 | 19.87 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.10.1 |
|
- Tokenizers 0.13.2 |
|
|