MixtureDensityNetworks v0.2.2
This release replaces the full covariance matrix in the MultivariateGMM layer with a diagonal covariance matrix.
Features
- This version replaces the full covariance matrix in the MultivariateGMM layer with a diagonal covariance matrix. This provides significantly faster training time at the cost of representation power. However, we can regain much of this power by increasing the number of mixture components in the predicted distribution.