mt5-base-finetuned-ar-fa
This model is a fine-tuned version of google/mt5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.0303
- Rouge-1: 26.73
- Rouge-2: 12.63
- Rouge-l: 23.96
- Gen Len: 18.99
- Bertscore: 71.41
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 |
|---|---|---|---|---|---|---|---|---|
| 3.7736 | 1.0 | 3287 | 3.2308 | 24.22 | 10.11 | 21.46 | 18.99 | 70.69 |
| 3.3783 | 2.0 | 6574 | 3.1283 | 25.28 | 10.9 | 22.43 | 18.99 | 71.02 |
| 3.2351 | 3.0 | 9861 | 3.0693 | 25.77 | 11.36 | 22.93 | 19.0 | 71.2 |
| 3.1363 | 4.0 | 13148 | 3.0421 | 25.88 | 11.57 | 23.08 | 18.99 | 71.22 |
| 3.0669 | 5.0 | 16435 | 3.0303 | 26.25 | 11.84 | 23.44 | 18.99 | 71.39 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1
- Downloads last month
- 10
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.