I have a data set which, no matter how I tune t-SNE, won't end in clearly separate clusters or even patterns and structures. Ultimately, it results in arbitrary distributed data points all over the plot with some more data points of the one class there and some of another one somewhere else.


Is it up to t-SNE, me and/or the data?

I'm using

                      , perplexity = 25
                      , max_iter = 1000000
                      , eta = 10
                      , check_duplicates = FALSE)

1 Answer 1


No, T-SNE does not have to result in clear clusters. It is a low dimension visualization of high dimension data. So, if you data points are well clustered in low dimension, it means that they can be classified in lower dimension. The idea behind T-SNE is to calculate probability of data points. Points far from each other have low probability. I would suggest to have a look at this link once, https://towardsdatascience.com/t-distributed-stochastic-neighbor-embedding-t-sne-bb60ff109561


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.