From: mblon...@gmail.com [mailto:mblon...@gmail.com] On Behalf Of
Mathieu Blondel
Sent: 18 October 2012 06:00
To: Didier Vila
Cc: scikit-learn-general@lists.sourceforge.net
Subject: Re: [Scikit-learn-general] HMM: Determination of the state
numbers
On Thu, Oct 18, 2012 at 1:37 AM, Didier
On Thu, Oct 18, 2012 at 1:37 AM, Didier Vila wrote:
> Problem 1: My problem is a risk problem at the moment.
>
> ** **
>
> **·**I want to represent the behaviour of my 20 000 time series
> and generates some monte carlo simulations using the method.sample()
>
> **·**Intuitiv
?
By Advance thanks for this.
Didier
From: Mathieu Blondel [mailto:math...@mblondel.org]
Sent: 17 October 2012 17:07
To: scikit-learn-general@lists.sourceforge.net
Subject: Re: [Scikit-learn-general] HMM: Determination of the state
numbers
On Thu, Oct 18, 2012 at 12:18 AM, Didier
On Thu, Oct 18, 2012 at 12:18 AM, Didier Vila wrote:
> Should I maximize the “score” function depending the number of state
> through several candidates that makes senses from a business point of view
> ?
>
The best criterion to optimize is not "score" but the evaluation metric you
really care
All,
I am using and testing the HMM module.
In particular, I would like to find a systematic way to define the
number of states.
As I have a a lot of time series ( ~20 000), I would like to make the
adjustment and optimization automatically.
Should I maximize the "score" functio
