Skip to main content

A library to work with complex number deep learning.

Project description

TF Complex Channels

TF complex channels contains a set of utility tools for working with complex numbers in a channel oriented way.

This approach is proposed and described in this blog post.

The primary purpose of this library and approach is to allow the community to push forward with academic research on complex number neural networks without waiting on ml library development.

Request for contributions

If an operation you want is not supported, please submit an issue or open a pull request

Basic Usage

import tf_complex_channels
print(x.dtype)
# > tf.complex64
print(x.shape)
# > (1, 2, 3)
x = complex_to_channels(x)
print(x.dtype)
# > tf.float64
print(x.shape)
# > (1, 2, 3, 2)

Documentation

Examples

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tf_complex_channels-0.0.1.tar.gz (1.6 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page