Data processing, testing, training, models, and predictors in here.
- will be updating the whole repo.
- more organization haha, this code was the result of a hackathon which is why its so hectic
- 2 different models
- one will take concantenated right hand and left hand melody data for input and output
- the other will take split seperate voices in the right hand and left hand melody data for multiple inputs and 1 output. (SATB format)
What it does:
- reads and stores music from music xml files
- processes data
- feeds into model
- takes output data and process into midi
Todo:
- clean up code
- work with more data, a lot more
- more efficient way to scrape data
- reorganize data processing
- get accuracy up up up!