mmiteva/qa_model_customs

mmiteva/qa_model-customs

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.3517
• Train End Logits Accuracy: 0.8772
• Train Start Logits Accuracy: 0.8735
• Validation Loss: 0.8793
• Validation End Logits Accuracy: 0.7642
• Validation Start Logits Accuracy: 0.7586
• Epoch: 4

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:

• optimizer: {'name': 'Adam', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 32050, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Train End Logits Accuracy	Train Start Logits Accuracy	Validation Loss	Validation End Logits Accuracy	Validation Start Logits Accuracy	Epoch
1.3795	0.6168	0.6015	0.9590	0.7074	0.6950	0
0.8193	0.7377	0.7260	0.8504	0.7313	0.7260	1
0.5982	0.8004	0.7932	0.8225	0.7505	0.7440	2
0.4467	0.8462	0.8405	0.8469	0.7633	0.7584	3
0.3517	0.8772	0.8735	0.8793	0.7642	0.7586	4

Framework versions

• Transformers 4.25.1
• TensorFlow 2.10.1
• Datasets 2.7.1
• Tokenizers 0.12.1