Viiksata/qa_model-davicni_800

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: 0.1075
• Train End Logits Accuracy: 0.9630
• Train Start Logits Accuracy: 0.9637
• Validation Loss: 0.5521
• Validation End Logits Accuracy: 0.8889
• Validation Start Logits Accuracy: 0.8925
• Epoch: 7

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', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 26320, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, '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
1.5290	0.6181	0.6265	0.7635	0.7831	0.7819	0
0.6245	0.8075	0.8159	0.5712	0.8379	0.8301	1
0.4064	0.8701	0.8721	0.5069	0.8656	0.8663	2
0.2854	0.9039	0.9096	0.4773	0.8813	0.8810	3
0.2145	0.9269	0.9287	0.4887	0.8732	0.8820	4
0.1669	0.9436	0.9462	0.4637	0.8938	0.8925	5
0.1300	0.9522	0.9563	0.5295	0.8938	0.8960	6
0.1075	0.9630	0.9637	0.5521	0.8889	0.8925	7

Framework versions

• Transformers 4.33.0
• TensorFlow 2.12.0
• Datasets 2.1.0
• Tokenizers 0.13.3