hubert-base-ls960-finetuned-gtzan

This model is a fine-tuned version of facebook/hubert-base-ls960 on the GTZAN dataset. It achieves the following results on the evaluation set:

• Loss: 0.6527
• Accuracy: 0.84

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1249	1.0	112	1.9377	0.43
1.6556	2.0	225	1.5867	0.47
1.2564	3.0	337	1.2670	0.56
1.0786	4.0	450	1.1080	0.59
0.895	5.0	562	0.8518	0.75
0.7177	6.0	675	1.0047	0.7
0.964	7.0	787	0.7430	0.75
0.4107	8.0	900	1.0347	0.71
0.4166	9.0	1012	0.5399	0.85
0.1234	10.0	1125	0.6266	0.83
0.0902	11.0	1237	0.6292	0.84
0.1211	12.0	1350	0.7393	0.84
0.4082	13.0	1462	0.6524	0.85
0.3442	14.0	1575	0.5732	0.86
0.0913	14.93	1680	0.6527	0.84

Framework versions

• Transformers 4.31.0.dev0
• Pytorch 1.13.0
• Datasets 2.1.0
• Tokenizers 0.13.3