"Module pytrees" that cleanly handle parameter trainability and transformations for JAX models.
-
Updated
May 11, 2023 - Python
#
bijector
Here are 2 public repositories matching this topic...
High-complexity implementation of Probabilistic Models (VAE & Normalizing Flows). Features dual-mode manifold discovery through $\beta$-regularization and Evidence Lower Bound (ELBO) maximization. Validated via FID (0.4473), utilizing TFP Bijectors and MultivariateNormalTriL layers.
python vae probabilistic-programming normalizing-flows variational-autoencoder generative-models imperial-college-london tensorflow-probability latent-space-interpolation bijector
-
Updated
Feb 2, 2026 - Jupyter Notebook
Add this topic to your repo
To associate your repository with the bijector topic, visit your repo's landing page and select "manage topics."