Learning-Wasserstein-Embeddings
Keras implementation of Deep Wasserstein Embeddings (Learning Wasserstein Embeddings) which approximates the Wasserstein distance with an autoencoder.
Prerequisite
Usage
First, compute the Wasserstein pairwise distance on n_pairwise couples of samples of the dataset of your choice. Previous records for mnist and quick draw are available in [data][./data]
$ python run_emd.py --dataset_name mnist --n_pairwise 10000
To train a siamese network to predict the pairwise wasserstein distance, use build_model.py :
$ python build_model.py --dataset_name mnist --embedding_size 50 --batch_size 32 --epochs 100
Examples
You can use your autoencoder to compute PCA, Barycenter Estimation or Interpolation with Wasserstein. The file test_model.py allows you to test your previously trained model to test on of these features or the MSE of your network, using the method_name attributes [MSE, PCA, BARYCENTER, INTERPOLATION].
$ python test_model.py --dataset_name mnist --method_name MSE