|
--- |
|
tags: |
|
- summarization |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: sinMT5-tuned |
|
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. --> |
|
|
|
# sinMT5-tuned |
|
|
|
This model is a fine-tuned version of [google/mT5](https://huggingface.co/csebuetnlp/mT5_multilingual_XLSum) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.8573 |
|
- Rouge1: 20.2531 |
|
- Rouge2: 8.1307 |
|
- Rougel: 19.3917 |
|
- Rougelsum: 20.0592 |
|
|
|
## 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: 0.00015652249866150822 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 7 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:------:|:-------:|:---------:| |
|
| 1.8651 | 1.0 | 1500 | 1.8070 | 17.676 | 7.1418 | 16.8638 | 17.457 | |
|
| 1.5527 | 2.0 | 3000 | 1.7804 | 21.1357 | 8.1386 | 20.122 | 20.8652 | |
|
| 1.3755 | 3.0 | 4500 | 1.7769 | 21.4151 | 8.5692 | 20.3204 | 21.1152 | |
|
| 1.2473 | 4.0 | 6000 | 1.7937 | 21.2434 | 8.2325 | 20.1332 | 21.0657 | |
|
| 1.1548 | 5.0 | 7500 | 1.8035 | 20.4298 | 8.2314 | 19.5909 | 20.2116 | |
|
| 1.0835 | 6.0 | 9000 | 1.8367 | 20.5427 | 8.2226 | 19.6134 | 20.2918 | |
|
| 1.0387 | 7.0 | 10500 | 1.8573 | 20.2531 | 8.1307 | 19.3917 | 20.0592 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.28.1 |
|
- Pytorch 2.0.0+cu118 |
|
- Tokenizers 0.13.3 |
|
|
