I have a follow up question regarding the usage of sample_weights for
fitting the RandomForestClassifier. Does the predict_proba method take the
sample weights (used during fitting) into account as well? I spent some
time trying to understand the _tree.pyc and tree.py files in the codebase
but stil
What do you need the random matrix for? A baseline?
A simple baseline is to predict the most common class for all obs.
On 15 Mar 2013 14:38, wrote:
> On Fri, Mar 15, 2013 at 10:24 AM, wrote:
> >> Having both margin fixed is an unlikely situation, especially for
> >> confusion matrices. Your ca
Hi
You are correct.
What do you mean by domain of applicability?
The class with the maximum probability gets the predicted discrete value
when you use fit ()
On 19 Mar 2013 14:44, wrote:
> Dear SciKitLearners,
>
> does anyone have experience in using RandomForest's predict_proba function
> as e
I'm experimenting with the examples/tutorials to get a feel for
scikit-learn.
"Faces recognition example using eigenfaces and SVMs" example.
Windows 7, python 2.73., etc. Installed all the packages to the latest
versions and the example was crashing here:
> face = np.asarray(imread(file_pat
Dear SciKitLearners,
does anyone have experience in using RandomForest's predict_proba function
as estimate for the domain of applicability?`
The situation is the following:
- data set contains 694 samples, each of which is defined by 94 features
- data has 2 classes: class0 and class1
- split i
>> II. Sometimes if attribute descriptions have multiple lines, a backtick
>> is needed at the end of continued lines. I still have no idea why and
>> what triggers this. Like I said, sometimes it's needed, sometimes it's
>> not.
> Backtick? You don't mean backslash?
Obviously yes :)
> This is no
On 03/19/2013 02:35 PM, Vlad Niculae wrote:
> I. In attributes, unlike in parameters, it's IMPORTANT to have one
Are you sure that this is not the case for parameters?
>
> Good: x : int
> Bad: x: int
This is done consistently for attributes as well as parameters (at least
in every PR I review ;)
W
Hello,
I apologize if this has already been discussed. I assume it hasn't,
and we should take a decision and write it down. Even if the codebase
isn't consistent, we should strive to have at least new PRs following
the rules.
A while back somebody asked me on IRC what the deal with backticks and
I think the below would suffice to interpolate precision (if I've
understood correctly). I'm not sure if there's a vectorised way to do it
given the existing implementation.
if interpolate:
for i in range(1, len(precision)):
precision[i] = precision[i-1:i+1].max()
Equivalently:
if in
On 03/19/2013 01:55 PM, Fimi wrote:
Hi Brian,
I will look into this paper in more detail. Thank you for your reply.
If I have to use opencv or other wrappers like it that hide SVM behind
its interface I will not use it. The purpose of this small project is
to learn Support Vector Machines.
Su
Hi Brian,
I will look into this paper in more detail. Thank you for your reply.
If I have to use opencv or other wrappers like it that hide SVM behind its
interface I will not use it. The purpose of this small project is to learn
Support Vector Machines.
Fimi
_
Hi Gilles,
Thank you very much for your answer.
I did really spend a lot more time on this then I had to. This is because the
topic is very interesting to me and I really went at it. I found out the hard
way that I could not use the whole image, it was bringing up other noisy items
into the
Sure, I could do that. Would need a couple days before I get to it,
though...
Regarding your statement that interpolated precision makes only sense on
ranked results, are you saying that whenever one is trying to build a
non-IR classifier then one should always go with ROC curves? I find myself
us
Hi Fimi.
Is there a reason you are not using the Viola-Jones implemented in OpenCV?
I should be available in SimpleCV, too, if you want a nice Python interface.
Cheers,
Andy
On 03/19/2013 05:19 AM, Fimi wrote:
Hello,
I've got non linear multiclass classification for support vector
machines to
As Gilles says, the scanning windows approach is pretty common for object
(and face) detection. Have you looked at the Viola Jones paper? It's the
standard for face detection and now that we have adaboost classifiers you
should be able to knock up an example quite quickly. Scikit Image might be
qui
Hi,
Short answer: you cant.
Longer answer: If you use as training samples the whole images (with faces
somewhere in there), then your model is learning to discriminate between
your 2 categories, from the whole images, with **no** information about
where the faces are actually located. As such, it
Sure, I could do that. Would need a couple days before I get to it,
though...
Regarding your statement that interpolated precision makes only sense on
ranked results, are you saying that whenever one is trying to build a
non-IR classifier then one should always go with ROC curves? I find myself
us
17 matches
Mail list logo
