whisper-tiny-finetuned-gtzan

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

• Loss: 0.7650
• Accuracy: 0.87

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 10
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0205	0.3274	37	1.6041	0.41
1.3349	0.6549	74	0.9462	0.67
1.1646	0.9823	111	0.9334	0.72
0.8737	1.3097	148	0.8974	0.64
0.8703	1.6372	185	0.7014	0.78
0.811	1.9646	222	0.8678	0.7
0.6429	2.2920	259	0.9130	0.66
0.6366	2.6195	296	0.7061	0.78
0.5858	2.9469	333	0.5549	0.82
0.3959	3.2743	370	0.5577	0.82
0.3343	3.6018	407	0.6203	0.83
0.3358	3.9292	444	0.8755	0.76
0.2574	4.2566	481	0.7690	0.79
0.1799	4.5841	518	0.7350	0.85
0.212	4.9115	555	0.6767	0.84
0.1553	5.2389	592	0.7819	0.84
0.1065	5.5664	629	0.9823	0.83
0.1151	5.8938	666	0.7709	0.84
0.0107	6.2212	703	0.7156	0.88
0.0564	6.5487	740	0.7283	0.88
0.0501	6.8761	777	0.7763	0.87
0.0846	7.2035	814	0.8221	0.83
0.0372	7.5310	851	0.7526	0.87
0.0015	7.8584	888	0.7705	0.87
0.0209	8.1858	925	0.7020	0.86
0.0114	8.5133	962	0.8043	0.86
0.0011	8.8407	999	0.7608	0.88
0.0018	9.1681	1036	0.7623	0.88
0.0009	9.4956	1073	0.7708	0.87
0.0219	9.8230	1110	0.7650	0.87

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1