Implement the pop-music transformer and integrate with hugging face API
Please follow the steps in REMI. https://github.com/YatingMusic/remi
You can change --dict_path to specify whether using chord or not.
python main.py --dict_path your-path-here
python main.py \
--is_continue 1 \
--continue_pth your-path-here
--dict_path should be same as training option.
python main.py \
--is_train 0 \
--prompt 0 \
--output_path your-path-here \
--model_path your-path-here
python main.py --is_train 0 \
--prompt_path your-path-here \
--output_path your-path-here \
--model_path your-path-here