0
$\begingroup$

So I have a set of 3000 features from which I would like to generate clusters. I passed my features through the T-SNE algorithm to reduce dimensionality to 2 features, and clusters are really visible in this reduced space. How would I go about figuring out a deterministic way to create this same T-SNE output to use for clustering?

$\endgroup$
  • 3
    $\begingroup$ Re-use the random number seed; e.g., using random_state in sklearn.manifold.TSNE. Welcome to the site. $\endgroup$ – Emre Apr 30 '18 at 20:32
2
$\begingroup$

Do not use tSNE visualizations for clustering. The results are misleading.

See this great answer: https://stats.stackexchange.com/a/264647/7828

Apart from that, you just need to fix the initial positions. For examples by fixing the random generator seed. But the fact thar it does not work every time should already warn you that it is not too reliable...

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.