-
There's been a lot of debate about whether we should incorporate sparse vectors into Julia 0.4.
Here are three options:
- (A) Use `SparseMatrixCSC` to emulate a sparse vector. This, as many have poin…
-
This is building on the discussions in #140 and #141. Sorry for the long text, but since this is a pretty important part of the framework, let's not leave room for misunderstanding!
I agree with http…
-
Lasagne `Layers` compute their outputs by calling `get_output()` recursively on their parents. When a network has a tree structure, which is usually the case, this works fine. But when a network has a…
-
We need more of matrix.hpp VS matrix_impl.hpp
@ghorn is of the opinion that even one liners should move tot impl..
-
I tried to build the scikit-learn docs for version 0.12 using Python 2.7.3 and Sphinx 1.1.3 on OSX 10.8 with the command `make html-noplot` but Sphinx fails with the following stacktrace:
``` python
…
-
Hi,
I think there are some things in https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/hmm.py that could be improved.
## _BaseHMM.algorithm handling
- _BaseHMM.decode has an 'algorithm…
kmike updated
10 years ago
-
The licensing in individual files is a bit all over the place. Currently, we have the following (each followed by a list of the files it occurs in).
License: BSD3
- ./sklearn/tree/tree.py:
License: …
-
In trying to implement an MPI backend, I've run up to the issue of where exactly to draw the line between front and back end. The current mechanism for choosing a backend is basically providing a diff…