Marmara-NLP/Turna_Question_answering

TURNA-finetuned_QA

This model is a fine-tuned version of boun-tabi-LMG/TURNA on an unknown dataset.

Model description

Turkish language model trained for question answering

Intended uses & limitations

This model can be used as a starting point to further train the Turna model for question answering, for now it's still not well trained and may require a larger dataset

Training and evaluation data

Clips/Mfaq turkish dataset: huggingface.co/datasets/clips/mfaq/viewer/tr_flat Size: 100k rows training, 2.7k validation This is a multilingual corpus of Frequently Asked Questions parsed from the Common Crawl.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1