DarijaBERT-finetuned-Arabic-SQuAD

This model is a fine-tuned version of SI2M-Lab/DarijaBERT on the Mostafa3zazi/Arabic_SQuAD dataset. It achieves the following results on the evaluation set:

• Loss: 4.7451

Model description

This model is a fine tuned version of DarijaBERT, specifically trained on the Arabic Squad dataset. It is designed to perform question answering in the Darija dialect, which is a colloquial Arabic variant spoken primarily in Morocco.

Intended uses & limitations

• the fine-tuning was done on arabic dataset so the results may not always be accurate escpacially with rare word in darija

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3