Gaurav Parmar, Dacheng Li, Kwonjoon Lee, Zhuowen Tu
CVPR 2021
Figure: Sampling Results of Re-implementation.
This folder provides a re-implementation of this paper in PyTorch, developed as part of the course METU CENG 796 - Deep Generative Models. The re-implementation is provided by:
- Aybora Köksal, [email protected]
- Halil Çağrı Bilgi, [email protected]
Please see the jupyter notebook file main.ipynb for a summary of paper, the implementation notes and our experimental results.
First anaconda package manager has to be installed on your system.
Then, to create the correct dependecies, run the below command.
conda env create --file requirements.txt
Note: This requirements txt is only for cpu use
Activate the conda environment
conda activate DC-VAE-env
To train the model use the below command. This command will start training, and creates a runs folder on the main directory where the metrics and logs of each experiment are easily tracktable.
python run.py
Produce qualitative and quantitative results with pre-trained model, look for the RESULTS part in main.ipynb