Bhtsne for large datasets

[aaronfderybel]

Hello @lvdmaaten ,

I've read on your tSNE homepage that you can handle datasets with up to 30 million examples https://lvdmaaten.github.io/tsne/. I'm currently working in google colab

I currently have a dataset with 2 million examples and each example is a 100-d vector.
Using verbose= False, I get the following:

Using verbose=True as suggested I get:

I'm not sure what this means or how i should proceed. The example with the Mnist dataset works perfect using verbose=False

[SamGG]

Not the solution, but for large dataset you should also look at
http://www.omiq.ai/opt-SNE
https://www.cytosplore.org
Best.