On 19 December 2011 04:16, Fernando Perez wrote:
> On Sun, Dec 18, 2011 at 4:08 AM, Gael Varoquaux
> wrote:
> > I suggest that we use the following conventions:
>
> Great idea. I really hope github adds soon support for proper labels
> and milestones to PRs (I've filed a ticket asking for it al
I see. Makes sense. Thanks guys.
On Mon, Dec 19, 2011 at 6:21 PM, Gael Varoquaux <
gael.varoqu...@normalesup.org> wrote:
> On Mon, Dec 19, 2011 at 10:24:18AM -0500, josef.p...@gmail.com wrote:
> > On Mon, Dec 19, 2011 at 10:19 AM, Jieyun Fu wrote:
> > > It's Wald's Z test in here. Search for Wal
On Mon, Dec 19, 2011 at 10:24:18AM -0500, josef.p...@gmail.com wrote:
> On Mon, Dec 19, 2011 at 10:19 AM, Jieyun Fu wrote:
> > It's Wald's Z test in here. Search for Wald Test in
> > http://userwww.sfsu.edu/~efc/classes/biol710/logistic/logisticreg.htm
> > In ordinary linear regression they like
2011/12/19 Olivier Grisel :
> Maybe we could move the fast_svd into the future
> sklearn.random_projection package and the graph_shortest_path in a
> sklearn.graph package that would also host the kneighbors_graph func?
I'm not sure if we should be developing a "machine learning & graph
algorithms
Good
I think for now I'll keep everything in sklearn/utils, but just clean
things up and write some documentation. I sent a message out to
scipy-dev floating the idea of adding graph_shortest_path to scipy:
we'll see where that goes.
Jake
Olivier Grisel wrote:
> I think I am overall +1 but I
I think I am overall +1 but I don't really like the
sklearn.enhancements package name. What functions would go in such
package?
- fast_svd
- graph_shortest_path
- what else?
Maybe we could move the fast_svd into the future
sklearn.random_projection package and the graph_shortest_path in a
sklearn
Someone suggested saving the figures in the examples as jpg to save room.
As far as I see it, this is not possible.
The pylab docstring says:
*format*:
One of the file extensions supported by the active
backend. Most backends support png, pdf, ps, eps and svg
And when I try
On Mon, Dec 19, 2011 at 10:19 AM, Jieyun Fu wrote:
> It's Wald's Z test in here. Search for Wald Test in
>
> http://userwww.sfsu.edu/~efc/classes/biol710/logistic/logisticreg.htm
>
> In ordinary linear regression they like to call it t-statistics. I guess
> it's just a terminology thing.
Since t
It's Wald's Z test in here. Search for Wald Test in
http://userwww.sfsu.edu/~efc/classes/biol710/logistic/logisticreg.htm
In ordinary linear regression they like to call it t-statistics. I guess
it's just a terminology thing.
Thanks again,
On Mon, Dec 19, 2011 at 10:14 AM, Alexandre Gramfort <
what is : \hat{ beta } / std.err ?
given a fitted model that gives clf.coef_ for the estimated coef?
Alex
On Mon, Dec 19, 2011 at 4:08 PM, Jieyun Fu wrote:
> Hi Alex,
>
> I am sorry to bother you again, but it doesn't look like that
>
> univariate_selection.f_classif
>
> gives me the t-statisti
Hi Alex,
I am sorry to bother you again, but it doesn't look like that
univariate_selection.f_classif
gives me the t-statistics of linear regression or logistic regression. It
gives F-value, but what I am looking for is \hat{ beta } / std.err
Thank you so much!
On Mon, Dec 19, 2011 at 9:47 AM,
> I haven't found your example. Can you post a link?
look at the docstring of cross_val_score
> I was looking
> at http://scikit-learn.org/stable/modules/classes.html#module-sklearn.metrics
>
> By the way, Is there a built-in method to calculate t-stat ? I couldn't find
> that even for ordinary l
Hi Alex,
I haven't found your example. Can you post a link?
I was looking at
http://scikit-learn.org/stable/modules/classes.html#module-sklearn.metrics
By the way, Is there a built-in method to calculate t-stat ? I couldn't
find that even for ordinary linear regression. I wouldn't mind writing m
OK, but if basics methods are likely to become more efficient than already
existing ones, they should be included in more general packages at some
point (i.e. fast_svd may end into scipy.linalg).
--
Virgile Fritsch
On Mon, Dec 19, 2011 at 1:57 PM, xinfan meng wrote:
> Good points. Sometimes pe
For those interested,
I think that one might be helpful (I alluded to it)
http://www.stat.washington.edu/raftery/Research/PDF/Fraley2007JClass.pdf
Bertrand
--
Learn Windows Azure Live! Tuesday, Dec 13, 2011
Microsoft is
Good points. Sometimes people just want to use the nuts and bolts instead
of a full-fledge machine.
On Mon, Dec 19, 2011 at 6:35 PM, Jacob VanderPlas <
vanderp...@astro.washington.edu> wrote:
> Hi all,
> I think we should re-think our utils model. Currently, it's a mish-mash
> of tools for devel
+1
- Ursprüngliche Mail -
Von: "Jacob VanderPlas"
An: scikit-learn-general@lists.sourceforge.net
Gesendet: Montag, 19. Dezember 2011 11:35:23
Betreff: [Scikit-learn-general] Utils upgrade
Hi all,
I think we should re-think our utils model. Currently, it's a mish-mash
of tools for devel
Hi all,
I think we should re-think our utils model. Currently, it's a mish-mash
of tools for development (e.g. array2d, logsum, etc.) and performance
code that's better than many other available python implementations
(e.g. fast_svd, graph_shortest_path, etc.). As far as I can tell, none
of t
Timmy,
I'm taking a closer look at your problem now. There's something very
strange in your adjacency matrix: a small set of points are the nearest
neighbors of virtually every other point. I think this is leading to
the singular weight matrices which cause the error. I visualized it
like th
19 matches
Mail list logo
