whisper-base.en-finetuned-gtzan

This model is a fine-tuned version of openai/whisper-base.en on the GTZAN dataset. It achieves the following results on the evaluation set:

• Loss: 0.6266
• Accuracy: 0.88

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:

• learning_rate: 5e-05
• train_batch_size: 12
• eval_batch_size: 12
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 18

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.7396	1.0	75	1.6061	0.56
0.8839	2.0	150	0.8286	0.77
0.7631	3.0	225	0.6353	0.81
0.4049	4.0	300	0.5840	0.82
0.3031	5.0	375	0.4069	0.88
0.3031	6.0	450	0.7152	0.81
0.2879	7.0	525	0.7061	0.85
0.0301	8.0	600	0.5691	0.89
0.0311	9.0	675	0.6153	0.88
0.0025	10.0	750	0.5463	0.88
0.0036	11.0	825	0.6017	0.89
0.0016	12.0	900	0.6859	0.85
0.0014	13.0	975	0.5887	0.89
0.0012	14.0	1050	0.6525	0.9
0.0011	15.0	1125	0.6289	0.89
0.0011	16.0	1200	0.6277	0.88
0.001	17.0	1275	0.6274	0.88
0.0611	18.0	1350	0.6266	0.88

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.3
• Tokenizers 0.13.3