| --- |
| tags: |
| - generated_from_trainer |
| metrics: |
| - rouge |
| model-index: |
| - name: bartpho-word-base |
| 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. --> |
|
|
| # bartpho-word-base |
|
|
| This model is a fine-tuned version of [vinai/bartpho-word-base](https://huggingface.co/vinai/bartpho-word-base) on the None dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 2.1067 |
| - Rouge1: 49.261 |
| - Rouge2: 22.7386 |
| - Rougel: 33.4517 |
| - Rougelsum: 35.6277 |
| - Gen Len: 19.8061 |
|
|
| ## 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: 2e-05 |
| - train_batch_size: 4 |
| - eval_batch_size: 4 |
| - seed: 42 |
| - gradient_accumulation_steps: 8 |
| - total_train_batch_size: 32 |
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| - lr_scheduler_type: linear |
| - num_epochs: 5 |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |
| |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:| |
| | 2.7155 | 1.0 | 309 | 2.1805 | 48.1588 | 21.5414 | 32.4169 | 34.6188 | 19.7453 | |
| | 2.2951 | 2.0 | 619 | 2.1331 | 48.5623 | 21.8348 | 32.7922 | 34.9789 | 19.7525 | |
| | 2.1676 | 3.0 | 929 | 2.1193 | 49.2433 | 22.8931 | 33.4674 | 35.738 | 19.7881 | |
| | 2.057 | 4.0 | 1239 | 2.1210 | 49.2825 | 22.7084 | 33.4667 | 35.5888 | 19.7759 | |
| | 2.0153 | 4.99 | 1545 | 2.1067 | 49.261 | 22.7386 | 33.4517 | 35.6277 | 19.8061 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.28.0 |
| - Pytorch 1.12.0+cu102 |
| - Datasets 2.12.0 |
| - Tokenizers 0.13.3 |
| |
