music_genres_classification-finetuned-gtzan

This model is a fine-tuned version of dima806/music_genres_classification on the GTZAN dataset. It achieves the following results on the evaluation set:

• Loss: 1.0028
• 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: 5e-05
• train_batch_size: 5
• eval_batch_size: 5
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.12
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5758	1.0	180	1.5756	0.52
1.079	2.0	360	1.2787	0.6
1.186	3.0	540	1.3863	0.58
0.9181	4.0	720	1.3967	0.64
0.4356	5.0	900	1.2449	0.67
0.4013	6.0	1080	1.2714	0.71
0.5518	7.0	1260	0.8282	0.8
0.4808	8.0	1440	1.3598	0.75
0.3608	9.0	1620	1.1908	0.8
0.181	10.0	1800	0.9824	0.83
0.0553	11.0	1980	1.0336	0.84
0.2445	12.0	2160	1.1085	0.83
0.0103	13.0	2340	1.1288	0.84
0.2437	14.0	2520	1.0183	0.85
0.0921	15.0	2700	1.0028	0.87

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2