Machine learning tools and framework for automatic music transcription
Project description
Automatic Music Transcription (AMT) Tools
Implements a customizable machine learning pipeline for AMT in PyTorch. This framework abstracts various elements of the AMT task, such as dataset usage, data formatting, feature extraction, model usage, output formatting, training, evaluation, and inference. This makes for easy modification and extension through inheritance.
The framework is a work-in-progress. Its development is ongoing to meet my evolving research needs.
Installation
Standard (PyPI)
Recommended for standard/quick usage
pip install amt-tools
Cloning Repository
Recommended for running examples or making experimental changes.
git clone https://github.com/cwitkowitz/amt-tools
pip install -e amt-tools
Usage
This repository can be used for many different purposes. Please see each subpackage for more information.
Additionally, several papers are implemented under the examples
subdirectory in standalone scripts which utilize the framework.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for amt_tools-0.1.4-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0e5c7f9f66250d5ea955ed5d33d6ba535c31c194b69e0c8a48fe4c4e2487d5c0 |
|
MD5 | 32c4813aa74ab40a98d90cc6943a42f8 |
|
BLAKE2b-256 | d1a487d43ff3c287bb784bc9b9f2b67e139584e95f85ed1fc77b130a62218067 |