An implementation of the mnist image generator using CNF.
(CNF is generative model which is introduced in NeuralODE.)
Get a latent of image using an image encoder and train CNF to generate latent.
During inference, generated latent is decoded by image decoder.
$t_0 = 0,\ t_1 = 10$ $\mathbf{z}_{t_0} \sim \mathcal{N}(0, I)$ $\mathbf{z}_{t_1}: target\ distribution$
(In the code, the covariance matrix of
Visualize the sample of
Without Encoder Condition Without Discriminator Latent Dimension = 2 | With Encoder Condition With Discriminator Latent Dimension = 8 |
Latent dimension is set to 2.
Latent Generated by ImageEncoder | Latent Generated by CNF |
python3 train.py