Tips to accelarate

[eoteroe]

Wondering if is possible to use threads. I was applying tsne to a dataset of size(50K,10) and it is really slow.
Can you give advice to handle big datasets?

[lejon]

Yes, unfortunately it is very slow. The current implementation is the original version of t-SNE and there is not much you can do there in terms of threads unfortunately. I have been working on a version of Barnes Hut t-SNE which is both much faster and parallelizable but I have not gotten it to work well yet unfortunately.As far as I know, if you need a fast version on large datasets you unfortunately have to go to another language. I have a Java version in another GitHub repo that is very fast, but there are other fast ones also. On 20 Jun 2024, at 22:20, Eliezer Otero Espinoza ***@***.***> wrote: Wondering if is possible to use threads. I was applying tsne to a dataset of size(50K,10) and it is really slow. Can you give advice to handle big datasets? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>