/VAEDemo

Some demo VAE models implemented with pytorch, including AE, DAE, VAE, CVAE.

Primary LanguagePython

VAE Demo

This repository includes some demo VAE models.

Note: The project refers to YixinChen-AI and timbmg

Datasets:

Models:

  • model1: AE

  • model2: DAE

  • model3: VAE

  • model4: CVAE

Unit Test

  • for loaders
PYTHONPATH=. python loaders/loader1.py
  • for modules
# AE
PYTHONPATH=. python modules/module1.py
# DAE
PYTHONPATH=. python modules/module2.py
# VAE
PYTHONPATH=. python modules/module3.py
# CVAE
PYTHONPATH=. python modules/module4.py

Main Process

python main.py

You can change the config either in the command line or in the file utils/parser.py

Here are the examples for each module:

# module1
python main.py \
    --name 1 \
    --module 1
# module2
python main.py \
    --name 2 \
    --module 2
# module3
python main.py \
    --name 3 \
    --module 3
# module4
python main.py \
    --name 4 \
    --module 4

Note (重点、难点、疑点、TODO、...)

1、目前 Encoder 和 Decoder 都是用 MLP 架构，对于 AE 和 DAE，当 latent_size 等于 2 时，降维效果并不理想，如果想要提升，可以使用卷积替代