This repository contains the codes of the following paper
BB Damodaran, B Kellenberger, R Flamary, D Tuia, N Courty, "DeepJDOT:Deep Joint Distribution Optimal Transport for Unsupervised Domain Adaptation", in European Conference on Computer Vision 2018 (ECCV-2018).
In order to run, the code requires the following Python modules:
- Numpy
- Matplotlib
- POT (Python Optimal Transport library)
- keras with tensorflow backend
- Deepjdot - module contains the implementation of the DeepJDOT
- dnn - import necessary functions from keras
- deepjdot_demo - DeepJDOT on the sample dataset
To run the DeepJDOT on the sample dataset, Please see or run the "deepjdot_demo.py"
To run on the real data set: SVHN --> MNIST, please see "deepjdot_svhn_mnist.py"