Metadata-Version: 2.1
Name: Mixture-Density-Nets
Version: 0.1.0
Summary: A small PyTorch library for Mixture Density Networks.
Home-page: https://github.com/romue404/mixture-density-nets
Author: Robert Müller
Author-email: robert.mueller1990@googlemail.com
License: MIT
Keywords: artificial intelligence,pytorch,mixture density network
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.6
Description-Content-Type: text/markdown
License-File: LICENSE

# 🎨 Mixture-Density-Nets
A small PyTorch library for Mixture Density Networks.

# Install
simply run 
``pip install mixture-density-nets``

# Example
```py
from mixture_density_nets import MDN, MDDistribution
# ....

mu, sigma, lambda_ = net(input_data)
dist = MDDistribution(mu, sigma, lambda_)
loss = dist.nll(targets).mean()

# ...
samples, clusters = dist.sample(n=20)  # draw 20 samples
```
