Hi Ali,
Can you describe the shapes/contents of those structures?
Am I right in thinking that this evaluates the entire tree for every
sample, rather than just the path from root to a single leaf? I can see
that as bringing speed gains if the process is vectorised over samples...?
- Joel
On We
Hi Arslan.
Have you tried the AdaBoost implementation in the current development
version?
Cheers,
Andy
On 07/24/2013 04:40 AM, Arslan, Ali wrote:
Hi,
I've been running adaboost with DecisionTreeClassifier in a for a
multiclass detection problem (comprises of multiple one-vs-all
problems). Th
Hi,
I've been running adaboost with DecisionTreeClassifier in a for a
multiclass detection problem (comprises of multiple one-vs-all problems).
The prediction method I'm using is like this:
for ii,thisLab in enumerate(allLearners):
res = np.zeros([dada.shape[0]], dtype='float16')
See ohe.active_features_. I agree it's not clearly documented (PR welcome),
but 'auto' encodes precisely those feature values seen in training (still
requiring non-negative integers), not merely their range.
On Wed, Jul 24, 2013 at 4:01 AM, Scott Alfeld wrote:
> Hi,
> When I fit a OneHotEncoder
Divisive clustering.
Intuitively, all points start in the same cluster. You then determine the
best way to split that cluster.
Recursively repeat until all points are in their own clusters.
http://nlp.stanford.edu/IR-book/html/htmledition/divisive-clustering-1.html
On 24 July 2013 10:01, Juan
On Wed, Jul 24, 2013 at 1:58 AM, Lars Buitinck wrote:
> And hierarchical isn't necessarily agglomerative. The alternative is
> something like HAClustering, which to me sounds like high-availability
> computer clusters.
>
Are you saying you could do a top-down hierarchy? I actually haven't seen
t
Dear scikit-learn team,
In fact, we have taken not only the logo but the entire template as a base
for nilearn website temporarily, but sometimes temporary things last a bit
long. The entire website is currently under refurbishment, but I've changed
the logo and switched some colors to avoid furth
On 23 July 2013 21:43, Andreas Mueller wrote:
> On 07/23/2013 07:39 PM, Nelle Varoquaux wrote:
> > Hello,
> >
> > After a discussion with Olivier and Gaël, we think it is best to go
> > with ``from matplotlib import pyplot as plt`` instead of ``import
> > pylab as pl``.
>
> +1 for matplotlib.pypl
On 07/23/2013 07:39 PM, Nelle Varoquaux wrote:
> Hello,
>
> After a discussion with Olivier and Gaël, we think it is best to go
> with ``from matplotlib import pyplot as plt`` instead of ``import
> pylab as pl``.
+1 for matplotlib.pyplot
I'm a bit surprised that
from matplotlib import pyplot a
Hi,
When I fit a OneHotEncoder, it sometimes encodes different values to the
same new vector, depending on wether n_values is explicit or 'auto'.
I wrote a brief script to demonstrate the issue, below.
Note that [[0, 1]] is either getting encoded as the same thing as [[0, 0]],
or as a different th
I was confused by installing matplotlib then importing pylab - this
clarifies it.
Regards,
Nigel Legg
07914 740972
http://www.trevanianlegg.co.uk
http://twitter.com/nigellegg
http://uk.linkedin.com/in/nigellegg
On 23 July 2013 18:52, Lars Buitinck wrote:
> 2013/7/23 Nelle Varoquaux :
> > 1. b
2013/7/23 Nelle Varoquaux :
> 1. beginners are confused on which python package to install when having an
> import error on ``import pylab as pl``;
Count me as a beginner, I thought pylab came from IPython instead of matplotlib.
--
Lars Buitinck
Scientific programmer, ILPS
University of Amsterda
2013/7/23 Nelle Varoquaux :
> Hello,
>
> After a discussion with Olivier and Gaël, we think it is best to go with
> ``from matplotlib import pyplot as plt`` instead of ``import pylab as pl``.
> This is implies minor modifications to the examples. I'll update the current
> examples. Please make sure
Hello,
After a discussion with Olivier and Gaël, we think it is best to go with
``from matplotlib import pyplot as plt`` instead of ``import pylab as pl``.
This is implies minor modifications to the examples. I'll update the
current examples. Please make sure when you create a new pull request to
2013/7/23 Juan Nunez-Iglesias :
> I'd vote for HierarchicalClustering, since, as Robert said, agglomerative is
> not necessarily hierarchical. Is Agglomerative really any more descriptive?
> That's not obvious to me.
And hierarchical isn't necessarily agglomerative. The alternative is
something li
I'd vote for HierarchicalClustering, since, as Robert said, agglomerative
is not necessarily hierarchical. Is Agglomerative really any more
descriptive? That's not obvious to me.
Also, the equivalent standard function in R is hclust, so that's something.
=)
On Tue, Jul 23, 2013 at 9:33 PM, Rober
I think the Python 2 + 3 support that we currently have is good enough
(with the help of the embedded copy of six).
I rather have explicit imports than from "future import *" because:
- it's not pep8
- it breaks static namespace analyzing tools such as pyflakes
--
Olivier
http://twitter.com/ogri
I'd go with AgglomerativeClusterer, Adding the hierarchical bit will make
it too long.
While agglomerative doesn't have to mean hierarchical, the usage is
consistent enough that I don't predict it causing much confusion.
On 23 July 2013 19:54, Andreas Mueller wrote:
> On 07/23/2013 10:52 AM, Al
Thanks Daniel.
On Tue, Jul 23, 2013 at 2:40 AM, Daniel Vainsencher <
daniel.vainsenc...@gmail.com> wrote:
> I opened the issue for the first, also mentioning the second problem.
>
> https://github.com/scikit-learn/scikit-learn/issues/2191
>
> Daniel
>
>
> On 07/22/2013 03:16 PM, Andreas Mueller
On 07/23/2013 10:44 AM, Gael Varoquaux wrote:
>> I'd rather raise a descriptive error message. I understand the
>> confusion from a Matlab user, but then I prefer enforcing Python
>> conventions over supporting Matlab conventions as well.
> +1 (you know, teaching people to fish instead of giving th
On 07/23/2013 10:52 AM, Alexandre Gramfort wrote:
> +1 for AgglomerativeClustering with linkage param to specify the criteria.
>
+1 if enough code is shared (which I think it will be)
--
See everything from the browser to
Easy to mistype but as appropriate as it gets. +1
On Tue, Jul 23, 2013 at 10:49 AM, Olivier Grisel
wrote:
> 2013/7/23 Lars Buitinck :
>> 2013/7/23 Olivier Grisel :
>>> 2013/7/23 Gael Varoquaux :
Hi people
How would you like an object that implements different hierarchical
aggl
+1 for AgglomerativeClustering with linkage param to specify the criteria.
Alex
On Tue, Jul 23, 2013 at 10:35 AM, Lars Buitinck wrote:
> 2013/7/23 Olivier Grisel :
>> 2013/7/23 Gael Varoquaux :
>>> Hi people
>>>
>>> How would you like an object that implements different hierarchical
>>> agglomer
2013/7/23 Lars Buitinck :
> 2013/7/23 Olivier Grisel :
>> 2013/7/23 Gael Varoquaux :
>>> Hi people
>>>
>>> How would you like an object that implements different hierarchical
>>> agglomerative clustering approaches be named?
>>>
>>> Options revolve around using the words 'linkage', as it is known i
> I'd rather raise a descriptive error message. I understand the
> confusion from a Matlab user, but then I prefer enforcing Python
> conventions over supporting Matlab conventions as well.
+1 (you know, teaching people to fish instead of giving them fish...)
G
--
2013/7/22 Andreas Mueller :
> I would really like all our estimators to accept (n_samples, 1) because
> that is indeed subtle and non-intuitive for beginners.
> Unfortunately that is not the case (yet).
I'd rather raise a descriptive error message. I understand the
confusion from a Matlab user, bu
2013/7/23 Olivier Grisel :
> 2013/7/23 Gael Varoquaux :
>> Hi people
>>
>> How would you like an object that implements different hierarchical
>> agglomerative clustering approaches be named?
>>
>> Options revolve around using the words 'linkage', as it is known in many
>> packages as such, 'hierar
2013/7/23 Gael Varoquaux :
> Hi people
>
> How would you like an object that implements different hierarchical
> agglomerative clustering approaches be named?
>
> Options revolve around using the words 'linkage', as it is known in many
> packages as such, 'hierarchical', as it is a fairly standard
+1 for linkage
Alex
On Tue, Jul 23, 2013 at 10:15 AM, Gael Varoquaux
wrote:
> Hi people
>
> How would you like an object that implements different hierarchical
> agglomerative clustering approaches be named?
>
> Options revolve around using the words 'linkage', as it is known in many
> packages
Hi people
How would you like an object that implements different hierarchical
agglomerative clustering approaches be named?
Options revolve around using the words 'linkage', as it is known in many
packages as such, 'hierarchical', as it is a fairly standard name for
these approaches.
Suggestions
30 matches
Mail list logo
