this isn't a neural network... I just need to calculate pairwise distances...
-
-
I have some x, y, points that's it.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I said KNN : k-nearest-neighbor Usually calculating distance is completely overkill with large data quantities.
-
explain...

-
usually we avoid any lasy learner with large/big data. It is simple, don't care about long training we want the fastest predictions
-
calculating distance is extremely intensive with high dimensions and big data. Try to avoid such learners as much as you can
-
I'm not doing any machine learning — just 2d points, euclidean distance. Or am I missing the point?
-
oh... I thought you was doing ML. Then just process data by chunk it's the easier.
-
is Googling "chunk numpy array" or similar the path to follow? Haven't done this before. I'm a(n ex) C coder.

-
I don't think you will find libraries for this. Just load the matrix little by little. Maybe this can help http://stackoverflow.com/questions/14551451/processing-a-very-very-big-data-set-in-python-memory-error …
- 12 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.