whisper_train_on_validated_cv_model__0010

This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.1998
• Train Accuracy: 0.0775
• Train Wermet: 2.6219
• Validation Loss: 0.3848
• Validation Accuracy: 0.0707
• Validation Wermet: 3.0529
• Epoch: 9

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': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Train Accuracy	Train Wermet	Validation Loss	Validation Accuracy	Validation Wermet	Epoch
2.2992	0.0331	6.3187	1.9677	0.0372	7.2782	0
1.7447	0.0431	5.6345	1.7865	0.0408	6.3969	1
1.5036	0.0480	5.6512	1.4821	0.0468	5.1810	2
1.0955	0.0567	4.4606	1.0492	0.0557	3.7886	3
0.7076	0.0654	4.9157	0.7312	0.0627	4.2416	4
0.4703	0.0710	4.9749	0.5668	0.0664	4.8883	5
0.3552	0.0737	5.1823	0.4725	0.0685	4.8141	6
0.2865	0.0754	4.1024	0.4358	0.0694	3.8550	7
0.2380	0.0765	3.3689	0.3947	0.0704	2.2953	8
0.1998	0.0775	2.6219	0.3848	0.0707	3.0529	9

Framework versions

• Transformers 4.32.0.dev0
• TensorFlow 2.12.0
• Tokenizers 0.13.3