In the Shogun library, the t-SNE algorithm is implemented in the CTDistributedStochasticNeighborEmbedding class. Objects of this class should be configured with the target number of dimensions and the set_target_dim() method. The fit() and transform() methods should be used for training and data dimensionality reduction, respectively:

 void TSNEReduction(Some<CDenseFeatures<DataType>> features,
                    const int target_dim) {
   auto tsne = some<CTDistributedStochasticNeighborEmbedding>();
   tsne->set_target_dim(target_dim);
   tsne->fit(features);
 
   auto new_features =
       static_cast<CDenseFeatures<DataType> *>(tsne->transform(features));
 
   auto feature_matrix = new_features->get_feature_matrix();
   for (index_t i = 0; i < new_features->get_num_vectors(); ++i) {
     auto new_vector = feature_matrix.get_column(i);
   }
 }

The following graph shows the result of applying the Shogun t-SNE implementation to our data: