salami_truncsplit_model

This model is a fine-tuned version of facebook/musicgen-melody on the TAUFIQSYED/SALAMI_CLEANED_SAMPLED - DEFAULT dataset. It achieves the following results on the evaluation set:

• Loss: 6.1330
• Clap: 0.1080

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: 0.0002
• train_batch_size: 1
• eval_batch_size: 1
• seed: 456
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Use adamw_torch with betas=(0.9,0.99) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 4.0

Training results

Training Loss	Epoch	Step	Validation Loss	Clap
6.4684	0.4813	25	6.0063	0.0988
6.4209	0.9627	50	6.0700	0.1396
6.7252	1.4428	75	6.0365	0.1031
6.1879	1.9242	100	5.8966	0.1233
6.3642	2.4043	125	6.8064	0.1065
6.2382	2.8857	150	6.1185	0.0766
6.3169	3.3658	175	6.1162	0.0969
5.7667	3.8472	200	6.1305	0.1143

Framework versions

• PEFT 0.13.2
• Transformers 4.47.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3