ModelWithMidiTok

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

• Loss: 8.2405

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: 2e-05
• 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
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	138	9.7069
No log	2.0	276	9.3817
No log	3.0	414	9.2525
9.5531	4.0	552	9.2097
9.5531	5.0	690	9.1881
9.5531	6.0	828	9.1586
9.5531	7.0	966	9.1142
9.1813	8.0	1104	9.0502
9.1813	9.0	1242	8.9741
9.1813	10.0	1380	8.8876
8.9874	11.0	1518	8.8039
8.9874	12.0	1656	8.7257
8.9874	13.0	1794	8.6524
8.9874	14.0	1932	8.5897
8.7156	15.0	2070	8.5348
8.7156	16.0	2208	8.4827
8.7156	17.0	2346	8.4419
8.7156	18.0	2484	8.4083
8.509	19.0	2622	8.3763
8.509	20.0	2760	8.3504
8.509	21.0	2898	8.3271
8.3922	22.0	3036	8.3093
8.3922	23.0	3174	8.2929
8.3922	24.0	3312	8.2786
8.3922	25.0	3450	8.2666
8.3214	26.0	3588	8.2577
8.3214	27.0	3726	8.2494
8.3214	28.0	3864	8.2447
8.2867	29.0	4002	8.2412
8.2867	30.0	4140	8.2405

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1