Hello,
Is there any way to detect matching features between 2 images in scikit
image or scikit learn. I am searching something like the SIFT or SURF
algorithms available in opencv.
I have seen the match_template() but it requires a template(region within
the image) and the source image. I need som
`sparse_std` isn't intended for CSR, and it isn't intended for public use,
really. It *will* work for CSR, but only if you swap the shape parameters,
after which it will give the stdev of the other axis.
On 25 March 2014 06:56, Lars Buitinck wrote:
> 2014-03-24 20:46 GMT+01:00 Manoj Kumar :
> >
You can just set `means_` and `covars_` attributes before fitting. But GMM
will also run the KMeans hotstarting for you if you like. e.g. if you don't
have a `means_` attribute and have and `m` in `init_params`, it runs kmeans
before fitting as shown on this line of the source code:
https://github.
2014-03-24 20:46 GMT+01:00 Manoj Kumar :
> It crashes always when n_samples != n_features for a CSR. Do you think this
> is a bug?
Retorical question? :) Seems like a pretty nasty bug to me. There's no
particular reason why sparse matrices should be square.
---
Well, I had a mistake.
It crashes always when n_samples != n_features for a CSR. Do you think this
is a bug?
On Tue, Mar 25, 2014 at 1:08 AM, Lars Buitinck wrote:
> 2014-03-24 20:05 GMT+01:00 Manoj Kumar :
> > X, y = make_regression(n_samples=100, n_features=2000)
>
> You're missing a random_s
2014-03-24 20:05 GMT+01:00 Manoj Kumar :
> X, y = make_regression(n_samples=100, n_features=2000)
You're missing a random_state arg here, which might explain why it
crashes sometimes but not always.
--
Learn Graph Databas
It looks like if sparse_std when n_samples != n_features for a CSC matrix,
then this segmentation error is raised. Shouldn't sparse_std be more
general? I've opened an issue here,
https://github.com/scikit-learn/scikit-learn/issues/3000.
P.S : Sorry for the noise.
On Tue, Mar 25, 2014 at 12:35 A
Minor mistake in the code,
X, y = make_regression(n_samples=100, n_features=2000)
X[X < 2.5] = 0.0
X = sparse.csr_matrix(X)
sparse_std(X.shape[0], X.shape[1], X.data, X.indices, X.indptr)
On Tue, Mar 25, 2014 at 12:32 AM, Manoj Kumar <
manojkumarsivaraj...@gmail.com> wrote:
> Hi,
>
> Running th
Hi,
Running the following code as a script gives me the following error,
Segmentation fault (core dumped). However if I run it in the interpreter
everything works (making me wonder how to debug it)
from scipy import sparse
from sklearn.datasets import make_regression
from sklearn.linear_model.bas
Quoting Arnaud Joly :
> Can you provide a gist of your code as to help you?
I have an implementation that mimics OnevsRestClassifier I want to
eventually try partial_fit since the number of samples is large. Here
is the rough outline.
#Tra
On Mon, Mar 24, 2014 at 04:05:32PM +0100, Nelle Varoquaux wrote:
> Does the test suit really takes more than 30 minutes on python 3? Are we able
> to reproduce this elsewhere?
It's super slow on jenkins.
G
--
Learn Graph
> > It's been a little while that our test suite doesn't finish under
> > jenkins with Python 3.3 because it takes too long to run:
> >
> https://jenkins.shiningpanda.com/scikit-learn/job/python-3.3-numpy-1.7.1-scipy-0.12.0/lastBuild/console
> >
> > We need to make sure that we keep a bound on the
2014-03-24 13:05 GMT+01:00 Gael Varoquaux :
> It's been a little while that our test suite doesn't finish under
> jenkins with Python 3.3 because it takes too long to run:
> https://jenkins.shiningpanda.com/scikit-learn/job/python-3.3-numpy-1.7.1-scipy-0.12.0/lastBuild/console
>
> We need to make s
It's been a little while that our test suite doesn't finish under
jenkins with Python 3.3 because it takes too long to run:
https://jenkins.shiningpanda.com/scikit-learn/job/python-3.3-numpy-1.7.1-scipy-0.12.0/lastBuild/console
We need to make sure that we keep a bound on the build time and test t
On Mon, Mar 24, 2014 at 06:54:45AM -0400, Hamzeh Alsalhi wrote:
> The changes I am making reside mostly in the Cython .pyx source code for the
> tree. I need to build and compile my changes to use in a small rf classifier
> script. So it seems like I need to go through the regular build process wit
Hi ,
I want to use GMM from sklearn for my application. Since
Gmm is sensitive to initial parameters, I derive initial
parameters from running K means on the data and
subsequently feed them to GMM.
How can I feed these parameters in GMM class of sklearn?
*params* ,
*init_params??*
--
W
Hi I have made a small contribution here
https://github.com/scikit-learn/scikit-learn/pull/2958
I am now working on prelimenary changes to implement
https://github.com/scikit-learn/scikit-learn/issues/655
The changes I am making reside mostly in the Cython .pyx source code for
the tree. I need
Hi,
Can you provide a gist of your code as to help you?
The pr 2458 isnt finished yet and there is possibly some quirk cases where
it might fail. However in the branch
https://github.com/arjoly/scikit-learn/commits/sparse-label_binarizer,
I almost finished the label binarizer part.
I can try t
18 matches
Mail list logo
