|
--- |
|
tags: |
|
- summarization |
|
- generated_from_trainer |
|
model-index: |
|
- name: led-risalah_data_v6 |
|
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. --> |
|
|
|
# led-risalah_data_v6 |
|
|
|
This model was trained from scratch on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.7883 |
|
- Rouge1 Precision: 0.7351 |
|
- Rouge1 Recall: 0.1832 |
|
- Rouge1 Fmeasure: 0.2927 |
|
|
|
## 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: 5.6e-05 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 Fmeasure | Rouge1 Precision | Rouge1 Recall | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------------:|:----------------:|:-------------:| |
|
| No log | 1.0 | 70 | 1.8476 | 0.2203 | 0.5271 | 0.1405 | |
|
| 1.9578 | 2.0 | 140 | 1.6793 | 0.2711 | 0.6467 | 0.1733 | |
|
| 1.3878 | 3.0 | 210 | 1.6549 | 0.261 | 0.6216 | 0.1666 | |
|
| 1.3878 | 4.0 | 280 | 1.7038 | 0.2814 | 0.656 | 0.1804 | |
|
| 1.1002 | 5.0 | 350 | 1.7134 | 0.2755 | 0.6577 | 0.1753 | |
|
| 0.8629 | 6.0 | 420 | 1.7733 | 0.2683 | 0.6436 | 0.1711 | |
|
| 0.8629 | 7.0 | 490 | 1.7755 | 0.2757 | 0.6489 | 0.1762 | |
|
| 0.6881 | 8.0 | 560 | 1.8848 | 0.2747 | 0.6502 | 0.1758 | |
|
| 0.5746 | 9.0 | 630 | 1.9318 | 0.2647 | 0.6305 | 0.1686 | |
|
| 0.479 | 10.0 | 700 | 1.9675 | 0.2655 | 0.633 | 0.1692 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.15.1 |
|
|
