In your case, you probably just perform a single pass through the data with
updateClassifier() (i.e., you perform a single epoch of training).
The Explorer will perform batch learning with 500 epochs by default:
The number of epochs to perform (batch learning only, default = 500)
If you want better results, you will need to iterate through the training data repeatedly
(unless you have a large amount of data relative to the difficulty of your learning
On 10 Nov 2015, at 23:23, Yanchao YU
Thanks for your replies. It is quite useful for my situation. Following the answer, I
re-built the whole SDG classifier and run it on my data set. However, I just found that
the result is worst than the model I build with a set of Instances.
And also I run the Weka 3.7 GUI on the same data set with the same settings. The results
is much better than the model I build with a set of Instances.
Could you tell me why it happens? please help me to solve this problem.
From: wekalist-bounces(a)list.waikato.ac.nz <wekalist-bounces(a)list.waikato.ac.nz> on
behalf of Mark Hall <mhall(a)waikato.ac.nz>
Sent: 09 November 2015 20:27
To: Weka machine learning workbench list.
Subject: Re: [Wekalist] Questions related to build an incremental Clasisfier
At a minimum, you need to know the structure of the training data in order to initialise
an incremental classifier in Weka. In other words, the buildClassifier() method needs to
be called once with an empty set of Instances (just attribute information defined). After
this, you can call updateClassifier() with individual instances.
From: <wekalist-bounces(a)list.waikato.ac.nz> on behalf of Yanchao YU
Reply-To: "Weka machine learning workbench list."
Date: Monday, 9 November 2015 10:09 pm
To: "Weka machine learning workbench list."
Subject: [Wekalist] Questions related to build an incremental Clasisfier
I am a beginner to use Weka on my project. I tried to implement an incremental Classifier
(SGD) for learning unknown objects. I followed some tutorials for building classifier
using existing data instances
). However, in my situation,
I cannot receive the training samples in the beginning. Is it possible to build classifier
without or with less instances?
My code is shown below:
_isExisted = true;
Thanks for your help in advance.
_______________________________________________ Wekalist mailing list Send posts
to:Wekalist@list.waikato.ac.nz List info and subscription status:
Wekalist mailing list
Send posts to: Wekalist(a)list.waikato.ac.nz
List info and subscription status: http://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html