Brandon Rose

@KennethNielsen that is interesting and might not be such an edge case for developers 😉. I won't have time to work on this until at least next week or the...

@loonies thanks for bumping this! @jmaupetit I agree on option 2. I'll take a look into fixing this next week. I'm glad to see that people are finding the feature...

@KennethNielsen that is awesome! I'm sorry to have fallen off this thread--things got busy but I am so glad you picked it up. Nice!

Hey @Shane-Neeley have you made any progress on this? If it's still broken I can look at your fork.

That was just laziness I guess. And to your point, I think my use case is somewhat an edge case: only proxy requests to Google, but do not proxy requests...

It's just for situations where for speed reasons I don't want to use different proxies for different tasks.

@bkieler as far as the specific question you can test this with a basic example: ``` from sklearn.metrics.pairwise import cosine_similarity ar1 = [0,3,4,1,3,5] ar2 = [1,2,4,3,1,3] print cosine_similarity(ar1,ar2) ``` returns...

Thanks @ouverz! 1. It sounds like you might have pretty tight clusters with a lot of similar words. When you conduct the reverse lookup using the pandas dataframe what is...

@ouverz sorry for the delay. In looking at your sample and the resultant clusters it looks like you have pretty homogenous documents which will have significant overlap. Your clusters are...

That does sound pretty intriguing. As for the number of features dropping when you increase `min_df`--that suggests to me that you have a significant number of features that occur in...