|
--- |
|
license: apache-2.0 |
|
tags: |
|
- summarization |
|
- arabic |
|
- ar |
|
- fa |
|
- persian |
|
- mt5 |
|
- Abstractive Summarization |
|
- generated_from_trainer |
|
model-index: |
|
- name: mt5-base-finetuned-arfa |
|
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-base-finetuned-arfa |
|
|
|
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: 3.1784 |
|
- Rouge-1: 25.68 |
|
- Rouge-2: 11.8 |
|
- Rouge-l: 22.99 |
|
- Gen Len: 18.99 |
|
- Bertscore: 71.78 |
|
|
|
## 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: 4 |
|
- label_smoothing_factor: 0.1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge-1 | Rouge-2 | Rouge-l | Gen Len | Bertscore | |
|
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:-------:|:---------:| |
|
| 3.9866 | 1.0 | 2649 | 3.3635 | 21.94 | 8.59 | 19.5 | 18.99 | 70.6 | |
|
| 3.5637 | 2.0 | 5298 | 3.2557 | 24.01 | 10.0 | 21.26 | 18.99 | 71.22 | |
|
| 3.4016 | 3.0 | 7947 | 3.2005 | 24.4 | 10.43 | 21.72 | 18.98 | 71.36 | |
|
| 3.2985 | 4.0 | 10596 | 3.1784 | 24.68 | 10.73 | 22.01 | 18.98 | 71.51 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.19.2 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.2.2 |
|
- Tokenizers 0.12.1 |
|
|