mmiteva/qa_model_test

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

• Train Loss: 0.4469
• Train End Logits Accuracy: 0.8470
• Train Start Logits Accuracy: 0.8386
• Validation Loss: 1.0938
• Validation End Logits Accuracy: 0.7318
• Validation Start Logits Accuracy: 0.7255
• 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': 108280, '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
1.4847	0.5934	0.5787	1.1159	0.6724	0.6590	0
0.9507	0.7042	0.6909	1.0094	0.6973	0.6875	1
0.7253	0.7637	0.7515	0.9841	0.7182	0.7124	2
0.5678	0.8090	0.7986	1.0107	0.7260	0.7194	3
0.4469	0.8470	0.8386	1.0938	0.7318	0.7255	4

Framework versions

• Transformers 4.20.1
• TensorFlow 2.9.2
• Datasets 2.1.0
• Tokenizers 0.12.1