mbert-bengali-tydiqa-qa-finetuned-RQA-factoid

This model is a fine-tuned version of sagorsarker/mbert-bengali-tydiqa-qa on the preprocessed only factoid based dataset BanglaRQA. It achieves the following results on the evaluation set:

• eval_loss: 0.7239
• eval_runtime: 34.5513
• eval_samples_per_second: 76.726
• eval_steps_per_second: 4.804
• epoch: 3.0
• step: 3999

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.19.1