It seems to me that the LSH forest is substituting for the `algorithm`
parameter, which selects between ball_tree, kd_tree and brute search for
nearest neighbour search. These are designed not to take additional
parameters.
So you need to accept additional parameters. You could indeed create
anoth
Hi,
I'm trying to use LSH Forest approximate neighbor search method to obtain
radius neighbors in DBSCAN. It adheres the API of sklearn.neighbors (at
least radius_neighbors method at this moment). But LSH Forest itself has a
set of parameters, so they need to be initialized.
I'm thinking about pa
I went through the documentation of the python wrapper for liblinear.
There are a few parameters such as
-p epsilon
-e epsilon (with -s)
-B bias
Assuming I'm using the default value of 1e-4 in sklearn, which values are used
for the parameters above. Does sklearn use default values, or different
I am using a simple text processing pipeline to perform sentiment
classification:
steps = [('vect', CountVectorizer()),
('tfidf', TfidfTransformer()),
('clf', LogisticRegression())]
pipe = Pipeline(steps)
With v0.15 the cross validation scores peak around 0.67 and with v0.15 they
peak at 0.55. T
it looks like you want convolutions... we have things for that
everyone has super optimized things for that
On Tue, Aug 5, 2014 at 10:57 AM, Bao Thien wrote:
> Thank you very much Yogesh :)
>
>
> On Tue, Aug 5, 2014 at 4:42 PM, Yogesh Karpate
> wrote:
>
>> Actually this question will b
Thank you very much Yogesh :)
On Tue, Aug 5, 2014 at 4:42 PM, Yogesh Karpate
wrote:
> Actually this question will be paid very less attention here because its
> ML forum . But anyways, you can try to
> vectorize your code using numpy or can use the same code snippet infused
> with
> numba (JI
Actually this question will be paid very less attention here because its
ML forum . But anyways, you can try to
vectorize your code using numpy or can use the same code snippet infused
with
numba (JIT Compiler). See this
http://nbviewer.ipython.org/github/aterrel/HPCPythonSC2012/blob/master/02_Sp
sorry, A is also array of (size1, size2)
On Tue, Aug 5, 2014 at 4:28 PM, Bao Thien wrote:
> Hi all scikit-learn users,
>
> Suppose that we have the code like:
>
>
> for i in np.arange(size1):
> for j in np.arange(size1):
> if j!=i:
> t = 0.
>
Hi all scikit-learn users,
Suppose that we have the code like:
for i in np.arange(size1):
for j in np.arange(size1):
if j!=i:
t = 0.
for k in np.arange(size2):
for l in np.arange(size2):
t = t + np.sum
Great, it works , thanks a lot.
But it has one little defect, after compilation, it reports that nosetests
cannot be found.
So I change your command to "make PYTHON=python3 NOSETESTS=nosetests3",
then it works perfectly :)
On Tue, Aug 5, 2014 at 8:58 PM, Lars Buitinck wrote:
> 2014-08-05 14:2
2014-08-05 14:27 GMT+02:00 Frank Dai :
> I want to compile scikit-learn under python 3.4, the following commands are
> what I'm doing:
>
> alias python=python3
Aliases only work in your current shell, not in make. Use
make PYTHON=python3
--
Hi all,
I want to compile scikit-learn under python 3.4, the following commands are
what I'm doing:
alias python=python3
sudo apt-get install python3-numpy python3-scipy python3-matplotlib
ipython3 ipython3-notebook python3-pandas python-sympy python3-nose
sudo apt-get install build-essential p
12 matches
Mail list logo
