|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: mt5-teste-full-length |
|
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-teste-full-length |
|
|
|
This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5750 |
|
- Rouge1: 0.4784 |
|
- Rouge2: 0.3008 |
|
- Rougel: 0.4185 |
|
- Rougelsum: 0.4212 |
|
|
|
## 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: 0.0005 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 16 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 90 |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:| |
|
| 2.9357 | 0.16 | 100 | 2.5583 | 0.2654 | 0.0431 | 0.1946 | 0.1951 | |
|
| 1.9974 | 0.33 | 200 | 1.7104 | 0.1803 | 0.0817 | 0.1712 | 0.1726 | |
|
| 1.4803 | 0.49 | 300 | 1.4404 | 0.1770 | 0.0695 | 0.1707 | 0.1727 | |
|
| 1.2432 | 0.65 | 400 | 1.0519 | 0.2809 | 0.1314 | 0.2509 | 0.2511 | |
|
| 0.8186 | 0.82 | 500 | 0.7386 | 0.3487 | 0.1767 | 0.2894 | 0.2903 | |
|
| 0.791 | 0.98 | 600 | 0.7135 | 0.3634 | 0.1912 | 0.3108 | 0.3108 | |
|
| 0.6697 | 1.15 | 700 | 0.6835 | 0.3874 | 0.1900 | 0.3123 | 0.3131 | |
|
| 0.7146 | 1.31 | 800 | 0.6657 | 0.3816 | 0.2209 | 0.3414 | 0.3428 | |
|
| 0.6957 | 1.47 | 900 | 0.6498 | 0.3878 | 0.2045 | 0.3336 | 0.3339 | |
|
| 0.6737 | 1.64 | 1000 | 0.6332 | 0.4094 | 0.2219 | 0.3524 | 0.3535 | |
|
| 0.6537 | 1.8 | 1100 | 0.6369 | 0.4401 | 0.2621 | 0.3629 | 0.3630 | |
|
| 0.6746 | 1.96 | 1200 | 0.6169 | 0.4369 | 0.2326 | 0.3566 | 0.3574 | |
|
| 0.5961 | 2.13 | 1300 | 0.6171 | 0.4364 | 0.2464 | 0.3666 | 0.3670 | |
|
| 0.5829 | 2.29 | 1400 | 0.6122 | 0.4539 | 0.2683 | 0.3813 | 0.3825 | |
|
| 0.6336 | 2.45 | 1500 | 0.5993 | 0.4347 | 0.2548 | 0.3660 | 0.3689 | |
|
| 0.5754 | 2.62 | 1600 | 0.5905 | 0.4575 | 0.2789 | 0.3856 | 0.3857 | |
|
| 0.5984 | 2.78 | 1700 | 0.5872 | 0.4630 | 0.2768 | 0.3915 | 0.3929 | |
|
| 0.5966 | 2.95 | 1800 | 0.5944 | 0.4605 | 0.2753 | 0.3822 | 0.3828 | |
|
| 0.5288 | 3.11 | 1900 | 0.5955 | 0.4520 | 0.2651 | 0.3874 | 0.3887 | |
|
| 0.5316 | 3.27 | 2000 | 0.5841 | 0.4649 | 0.2820 | 0.4052 | 0.4056 | |
|
| 0.5332 | 3.44 | 2100 | 0.5765 | 0.4861 | 0.3046 | 0.4021 | 0.4050 | |
|
| 0.5296 | 3.6 | 2200 | 0.5812 | 0.4610 | 0.2815 | 0.3976 | 0.4021 | |
|
| 0.5215 | 3.76 | 2300 | 0.5757 | 0.4724 | 0.2947 | 0.4122 | 0.4164 | |
|
| 0.5399 | 3.93 | 2400 | 0.5750 | 0.4784 | 0.3008 | 0.4185 | 0.4212 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.27.4 |
|
- Pytorch 1.13.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.13.2 |
|
|