ahmeddbahaa's picture
update model card README.md
0f8a1bc
---
tags:
- summarization
- persian
- MBart50
- Abstractive Summarization
- generated_from_trainer
datasets:
- xlsum
model-index:
- name: mbart-large-50-finetuned-persian
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. -->
# mbart-large-50-finetuned-persian
This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on the xlsum dataset.
It achieves the following results on the evaluation set:
- Loss: 4.1932
- Rouge-1: 26.11
- Rouge-2: 8.11
- Rouge-l: 21.09
- Gen Len: 37.29
- Bertscore: 71.08
## 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: 0.0005
- 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
- label_smoothing_factor: 0.1
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge-1 | Rouge-2 | Rouge-l | Gen Len | Bertscore |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:-------:|:---------:|
| 5.5612 | 1.0 | 1476 | 4.5015 | 17.07 | 3.14 | 13.54 | 47.49 | 66.83 |
| 4.3049 | 2.0 | 2952 | 4.1055 | 22.63 | 5.89 | 18.03 | 40.43 | 69.23 |
| 3.8154 | 3.0 | 4428 | 3.9822 | 24.57 | 7.15 | 19.74 | 37.35 | 70.36 |
| 3.3401 | 4.0 | 5904 | 4.0088 | 25.84 | 7.96 | 20.95 | 37.56 | 70.83 |
| 2.8879 | 5.0 | 7380 | 4.1932 | 26.24 | 8.26 | 21.23 | 37.78 | 71.05 |
### Framework versions
- Transformers 4.19.1
- Pytorch 1.11.0+cu113
- Datasets 2.2.1
- Tokenizers 0.12.1