Variational Autoencoder in TensorFlow

Package Dependencies

This repository trains the Variational Autoencoder in Tensorflow framework on Fashion-MNIST and Cartoon dataset. The cartoon dataset can be download from here.

The code is tested with:

• Cuda-11.1
• Cudnn-8.0

The Tensorflow notebook requires numpy, tf-nightly-gpu, opencv, sklearn. To get the versions of these packages you need for the program, use pip: (Make sure pip is upgraded: python3 -m pip install -U pip)

pip3 install -r requirements.txt 

Add Virtualenv as Python Kernel in Jupyterlab

• Activate the virtualenv

$ source your-venv/bin/activate

• Add the virtualenv as a jupyter kernel

(your-venv)$ ipython kernel install --name "local-venv" --user

Replace local-venv with your virtualenv name.