shobon latent space using GPLVM

Variational Autoencoders implementation in Keras.

Implemented the Variational AutoEnoder and Decoder

Implementation of my B.Sc. Thesis titled "MddGAN : Multilinear Analysis of the GAN Latent Space".

Extension of GANspace: https://github.com/harskish/ganspace

Exploring Latent Space via VAE

Pre-computed latent codes for controlling facial weight in StyleGAN and StyleGAN2 generated or projected images.

TensorFlow implementation of a simple Variational Autoencoder. Includes a simple GUI for exploring the learned latent space.

Space Model framework that allows for maintaining generalizability, and enhances the performance on the downstream task by utilizing task-specific context attribution. It is an external LLM layer, that improves accuracy in classification task for multiple datasets, such as HateXplain, IMDB movies reviews and more.

Source code and supplementary material for DIRAC: Diffusion-Based Representation Learning for Modality-Agnostic Compositionality

Pytorch + HuggingFace Diffusers project that experiments with latent diffusion model by training variational autoencoder (VAE) and a conditional U-Net.

Witchcraft is a toolkit capable of encoding documents of various content types and structure (flat, hierarchical, or series) into searchable latent vector space.

Une Expérience de Navigation dans le Latent Space

World Structured SwiGLU FFN

Toy example for a Conditional Variational Autoencoder in Keras. Regresses features from automatically generated images. Useful for learning about the concept.

Latent space explorer from a convolutional autoencoder trained on capital letters

Multi-Operational Mathematical Derivations in Latent Space

Deep Neural Net For Finding Similar Images With Hyperparameter Optimization + AWS And Azure GPU Capabilities

Exploring the use of Adversarial Constrained Autoencoder Interpolation (ACAI) to improve the quality of latent space for 3D human pose representation using the h36m dataset.