Base-Models-Finetuned-on-PersianQuAD
Collection
Base models finetuned with PersianQuAD and baseline dataset PersianQuAD. • 11 items • Updated
# Load model directly
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
tokenizer = AutoTokenizer.from_pretrained("Z-Jafari/bert-fa-base-uncased-finetuned-PersianQuAD-3epochs")
model = AutoModelForQuestionAnswering.from_pretrained("Z-Jafari/bert-fa-base-uncased-finetuned-PersianQuAD-3epochs")Quick Links
bert-fa-base-uncased-finetuned-PersianQuAD-3epochs
This model is a fine-tuned version of HooshvareLab/bert-fa-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2230
- {'exact_match': 76.8, 'f1': 80.59308322465655}
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.123 | 1.0 | 1291 | 1.0576 |
| 0.6574 | 2.0 | 2582 | 1.1094 |
| 0.4112 | 3.0 | 3873 | 1.2230 |
Framework versions
- Transformers 4.57.3
- Pytorch 2.9.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1
- Downloads last month
- 4
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("question-answering", model="Z-Jafari/bert-fa-base-uncased-finetuned-PersianQuAD-3epochs")