# How to recreate T-SNE dimensions deterministically?

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?

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