Stack Exchange network consists of 183 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.
Essentially what I could do is to set self.error = error at the end of _tsne(), in order to retrieve it from the TSNE instance afterwards. Yes, but that would mean changing sklearn.manifold code, and I was wondering if the developers thought of some other ways to get the information or if not why they didn't (i.e.: is 'error' considered useless by them?). Furthermore, if I changed that code I would need all the people running my code to have the same hack on their sklearn installations. Is that what you suggest, or did I get it wrong?
