Hey.
You can use expon:
http://nbviewer.ipython.org/github/amueller/pydata-nyc-advanced-sklearn/blob/master/Chapter%203%20-%20Randomized%20Hyper%20Parameter%20Search.ipynb
(it has a scale parameter)
Andy
On 04/07/2015 05:16 PM, Arnaud Fouchet wrote:
Hello,
I have a little question about Rando
Hello,
I have a little question about Random Search vs Grid Search. I am used to
grid search over parameter increasing exponentially (lambda = 0.01, 0.1,
... 1000). When you use randomized search, what type of distribution do you
use ? If I use an uniform distribution, I would test hyper-parameter
I talks about it a bunch here:
https://www.youtube.com/watch?v=0wUF_Ov8b0A
The paper is here: http://www.jmlr.org/papers/v13/bergstra12a.html
I'd be interested in the blog post.
Really I don't think searching over losses and regularization will give
you much usually.
I have not really seen any
Hi Roberto,
I'm no expert by any means, but I was reading a blog post the other day that
talked about using Random Search vs Grid Search. The gist of the article is
that, since you can feed distributions to Random Search and it selects values
randomly over the number of iterations you choose,
t;> From: Andreas Mueller [t3k...@gmail.com <mailto:t3k...@gmail.com>]
>> Sent: Tuesday, April 07, 2015 1:49 PM
>> To: scikit-learn-general@lists.sourceforge.net
>> <mailto:scikit-learn-general@lists.sourceforge.net>
>> Subject: Re: [Scikit-learn-general] CV with SVM
>>
would not prevent that
*From:* Andreas Mueller [t3k...@gmail.com]
*Sent:* Tuesday, April 07, 2015 1:49 PM
*To:* scikit-learn-general@lists.sourceforge.net
*Subject:* Re: [Scikit-learn-general] CV with SVM
No, it errors on
-learn-general@lists.sourceforge.net
Subject: Re: [Scikit-learn-general] CV with SVM
No, it errors on invalid configurations. You can specify lists of dictionaries
if you don't want all possible combinations.
There is not really any point in searching over "dual=True/False". Why would
No, it errors on invalid configurations. You can specify lists of
dictionaries if you don't want all possible combinations.
There is not really any point in searching over "dual=True/False". Why
would you want that?
On 04/07/2015 12:24 PM, Pagliari, Roberto wrote:
not all combinations of cost/l
