In the article Comprehensive Guide on t-SNE algorithm with implementation in R & Python.
It is mentioned that tsne is roboust to chagnes in perplexity hyperparameter.
but below article from sklearn has plots of different perplexity.
t-SNE: The effect of various perplexity values on the shape — scikit-learn 0.24.1 documentation
Which one is correct here?
1 post - 1 participant