Implemented Variational Autoencoder generative model in Keras for image generation and its latent space visualization on MNIST and CIFAR10 datasets

Semi-supervised learning with mnist using variational autoencoders. An unsupervised representation is learned which allows for superior classification results with limited labels.

Pytorch implementation of a Variational Autoencoder (VAE) that learns from the MNIST dataset and generates images of altered handwritten digits.

Stacked Denoising and Variational Autoencoder implementation for MNIST dataset

Implementation of CoordConv (Convolution and Deconvolution) for a Variational Autoencoder applied to MNIST

Exploring the depths of generative learning with a $\beta$-Variational Autoencoder ($\beta$-VAE) applied to the MNIST dataset for robust digit reconstruction and latent space analysis.

Implementation and analysis of a Conditional Variational Autoencoder (CVAE) in PyTorch for generating MNIST digits. Includes visualization of the 2D latent space using t-SNE and reconstruction quality assessment via PSNR/SSIM.

Tensorflow implementation of Variational Autoencoder for MNIST

A Variational Autoencoder (VAE) project designed to generate and reconstruct MNIST handwritten digit data. This repository demonstrates the use of deep learning techniques to model latent spaces and produce high-quality digit samples, showcasing the power of unsupervised learning for generative tasks.

Pytorch implementation of a Gaussian Mixture Variational Autoencoder used for clustering of the MNIST data set.

This repo explores Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs). It implements Control VAE for disentangled representation learning and trains various GAN models (Basic, WGAN, SS-GAN) to generate new data (dSprites, MNIST-like digits).

This paper focuses on Implemention of Deep Convolutional-Generative Adversarial Network and Variational Autoencoder for MNIST dataset consisting of 60000 records and for ANIME dataset of 80000 records and analyze the performance of both the models on these datasets

This repository contains implementations and experiments with various types of Variational Autoencoders (VAEs) applied to the MNIST dataset. VAEs are generative models that learn to encode and decode data in a probabilistic manner, making them particularly useful for image generation tasks.

No description available

No description available

An Implementation of Conditional Variational Autoencoders for MNIST.

Implementation of Variational Autoencoder in Pytorch for MNIST dataset.

A Pyro implementation of probabilistic variational autoencoders for MNIST.

Building a Variational Autoencoder (VAE) functionality for MNIST Dataset

Implementation of a variational autoencoder for generating MNIST samples

Variational Autoencoder (VAE) for MNIST using LSTM encoder and CNN decoder

Conditional Variational Autoencoder for MNIST with improved stability and digit-specific generation.

Deep Generative Models: Class-conditioned Variational Autoencoder (VAE) for generating MNIST digits.

This repository contains Autoencoders, Variational Autoencoders and GANS-Unsupervised Models developed for MNIST Dataset in Tensorflow and PyTorch.

Variational Autoencoder using the MNIST dataset. Also included, is an ANN and CNN for MNIST as well. Coded in Python, uses PyTorch

Evaluating custom architectures of autoencoders and variational autoencoders built on a custom backbone of ResNet blocks for denoising MNIST dataset images injected with gaussian noise

A Conditional Variational AutoEncoder (CVAE) for MNIST digit generation. Generate specific digits (0-9) on demand using one-hot encoded labels and raw MNIST data loading with PyTorch.

An interactive MNIST digit generator using a Conditional Variational Autoencoder (CVAE) trained with PyTorch and deployed via Streamlit for real-time digit synthesis.

Explorations in AI: Implementing Random Forest for price prediction, Variational Autoencoders for anomaly detection on Fashion MNIST, and integrating Llama-3 via Groq for NLP tasks.

The project aims to explore the potential of Conditional Variational Autoencoders (C-VAEs) for generating synthetic data from specific image classes in the MNIST dataset.