# midi-rae


<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->


              ▬▬    ▬▬▬    ▬▬                                       ▬▬                                  "Scrawl Me Maybe"   
                     ▬▬                                                                                                     
     ▬▬  ▬▬  ▬▬▬     ▬▬   ▬▬▬        ▬▬▬▬▬▬ ▬▬▬▬   ▬▬▬▬            ▬▬▬   ▬▬▬▬ ▬▬▬▬▬▬  ▬▬▬▬           ▬▬▬▬▬   ▬▬▬▬  ▬▬▬▬▬    
     ▬▬▬▬▬▬▬  ▬▬  ▬▬▬▬▬    ▬▬   ▬▬▬▬  ▬▬  ▬▬   ▬▬ ▬▬  ▬▬   ▬▬▬▬     ▬▬  ▬▬  ▬▬ ▬▬  ▬▬    ▬▬   ▬▬▬▬  ▬▬      ▬▬  ▬▬ ▬▬  ▬▬ 
     ▬▬ ▬ ▬▬  ▬▬ ▬▬  ▬▬    ▬▬         ▬▬    ▬▬▬▬▬ ▬▬▬▬▬▬            ▬▬  ▬▬▬▬▬▬ ▬▬  ▬▬ ▬▬▬▬▬          ▬▬▬▬▬  ▬▬  ▬▬ ▬▬  ▬▬   
     ▬▬   ▬▬  ▬▬ ▬▬  ▬▬    ▬▬         ▬▬   ▬▬  ▬▬ ▬▬            ▬▬  ▬▬  ▬▬     ▬▬▬▬▬ ▬▬  ▬▬              ▬▬ ▬▬  ▬▬ ▬▬  ▬▬ 
     ▬▬   ▬▬ ▬▬▬▬ ▬▬▬▬▬▬  ▬▬▬▬       ▬▬▬▬   ▬▬▬▬▬  ▬▬▬▬▬        ▬▬  ▬▬   ▬▬▬▬▬ ▬▬     ▬▬▬▬▬         ▬▬▬▬▬▬   ▬▬▬▬  ▬▬  ▬▬ 
                                                                 ▬▬▬▬         ▬▬▬▬      

TODO: Not documented yet because it’s not ready for anyone besides the
author to use.

## Developer Guide

If you are new to using `nbdev` here are some useful pointers to get you
started.

### Install midi_rae in Development mode

``` sh
# make sure midi_rae package is installed in development mode
$ pip install -e .

# make changes under nbs/ directory
# ...

# compile to have changes apply to midi_rae
$ nbdev_prepare
```

## Usage

### Installation

Install latest from the GitHub
[repository](https://github.com/drscotthawley/midi-rae):

``` sh
$ pip install git+https://github.com/drscotthawley/midi-rae.git
```

or from [pypi](https://pypi.org/project/midi-rae/):

``` sh
$ pip install midi-rae
```

### GPU Acceleration (Optional)

For faster UMAP and PCA projections using NVIDIA GPUs, install with the
`gpu` extra:

``` sh
$ pip install midi-rae[gpu]
```

This requires [RAPIDS](https://rapids.ai/) cuML and cuPy. See the
[RAPIDS installation guide](https://docs.rapids.ai/install) for setup
instructions — pip wheels are available for supported CUDA versions:

``` sh
$ pip install cuml-cu12 cupy-cuda12x  # for CUDA 12
```

Without GPU dependencies, the package falls back to CPU-based sklearn
and umap-learn automatically.

### Documentation

Documentation can be found hosted on this GitHub
[repository](https://github.com/drscotthawley/midi-rae)’s
[pages](https://drscotthawley.github.io/midi-rae/). Additionally you can
find package manager specific guidelines on
[conda](https://anaconda.org/drscotthawley/midi-rae) and
[pypi](https://pypi.org/project/midi-rae/) respectively.

## How to use

Fill me in please! Don’t forget code examples:

``` python
1+1
```

    2