CannyLab/tsne-cuda

How to include labels during embedding

meghbhalerao opened this issue · 1 comments

Hi all,
I have a very basic question. How do I incorporate the labels (for colouring the tsne plot) after the embedding has been learnt in this step: X_embedded = TSNE(n_components=2, perplexity=15, learning_rate=10).fit_transform(X). From what I understand, I see only X which are the high dimensional vectors.
Thanks,
Megh

rmrao commented

tSNE does not incorporate label information, it is simply a dimensionality reduction algorithm (not a clustering algorithm). The labels are for visualization purposes only and are passed in as colors when creating a scatterplot in matplotlib.