musicgen-electronic

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

musicgen-electronic

This model is a fine-tuned version of facebook/musicgen-small on the TruongScotl/electronic_music_generating dataset. For code (create data and training) you can check here: https://www.kaggle.com/code/truongscotl/fine-tuning-musicgen-with-lora

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: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.98,0.99) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

• epoch = 9.9487
• total_flos = 1509258GF
• train_loss = 6.414
• train_runtime = 3:20:31.12
• train_samples = 1559
• train_samples_per_second = 1.296
• train_steps_per_second = 0.081

Framework versions

• PEFT 0.12.0
• Transformers 4.42.3
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1