praca-magisterska/project
Cezary Pukownik [Impakt S.A.] eaaad76a2a added shell interactions
2019-10-25 11:23:59 +02:00
..
__pycache__ init commit 2019-10-24 14:01:43 +02:00
extract.py added shell interactions 2019-10-25 11:23:59 +02:00
generate.py added shell interactions 2019-10-25 11:23:59 +02:00
midi_processing.py init commit 2019-10-24 14:01:43 +02:00
model.py init commit 2019-10-24 14:01:43 +02:00
readme.md init commit 2019-10-24 14:01:43 +02:00
train.py init commit 2019-10-24 14:01:43 +02:00

MUSIC GENERATION USING DEEP LEARNING

AUTHOR: CEZARY PUKOWNIK

How to use:

  1. In folder ./midi_packs make folder with midi files you want train on
  2. Use extract.py to export data from midis

./extract.py [str: midi_pack_name] [str: name_of_session] --b [int: seq_len] -a [analize data first]

  1. Use train.py to train model

./train.py [str: name_of_session] --b [int: batch_size] --l [int: latent_space] --e [int: epochs] --i [str: instrument] -r [reset]

  1. Use generate.py to generate music from models

./generate.py [str: name_of_session] --n [number of generations] --m [mode {'from_seq','from_state'}]