I installed weka recently for a class in machine learning that I am taking and have been told that certain datasets are available in the weka folder called data. I installed the mac version of weka found at this link but I don't seem to have the sample datasets nor can I find them when I search on my mac through spotlight search. Does anyone know the location that mac would have them. Or does anyone have a link to the sample data folder online that I can download.
Not a Mac user, but you can find the datasets in the Weka git repository:
https://git.cms.waikato.ac.nz/weka/weka/-/tree/main/trunk/wekadocs/data
Or you can download them (and more) from sourceforge.net:
https://sourceforge.net/projects/weka/files/datasets/
The UCI datasets, which make up a large portion of the Weka example datasets, are available from this location:
https://sourceforge.net/projects/weka/files/datasets/UCI%20and%20StatLib/
For macOS, you can find the dataset files by right-click Weka in "Applications", click "Show Package Contents" then within the directory Contents/app/data.
Example, on my mac the location is:
./System/Volumes/Data/Applications/weka-3.8.6.app/Contents/app/data/
Related
I have a problem with SAS Enterprise Guide (EG) on my MacBook M1. I run Windows 11 on the machine using Parallel Desktop.
I got no problem to use the Data Integration (DI) and Forecast Studio (FS), I can work on both tools for my project.
But on EG, when I try to open the Libraries from server, I will get error "Unable to load the children for the item named 'Libraries'".
I search on Google and found below
http://support.sas.com/kb/65/222.html
It basically said that is due to permission on desktop and documents folder. But I can't any where to do the setup.
May I know anyone encounter this problem before and is there any solution?
According to SAS, EG is not support ARM processor. So can't use EG on M1.
I am following page for building chromedriver.exe from source code:
https://chromium.googlesource.com/chromium/src/+/master/docs/windows_build_instructions.md#Get-the-code
I have installed Visual Studio 2017 as stated.
Also downloaded and extracted depot_tools in C:\src\depot_tools.
I have installed Python 2 and msysgit previously.
I have added depot_tools in PATH System Variables.
I have also added 'DEPOT_TOOLS_WIN_TOOLCHAIN' variable in System Variables with value 0.
Now as stated I am running gclient command from cmd but nothing is working.
I can't move further.
Also if you have any Video Referral please share as I didn't get any on Google search.
And if you know it well and would like to help please share your email id or twitter handle because it seems hard to get answers to such queries as there are only few such questions on StackOverflow and that too uanswered. So a contact might be helpful.
I am getting a different error from this post (KNIME 3.2 Tableau Node error). I believe my problem precedes his. I suspect my problem is in setting up my dependencies and PATH variables.
I have downloaded the Tableau SDK, and placed it both in the KNIME Workspace, and in Tableau's program files. I have configured the linked resources page of the KNIME to point to each file individually and to the bin directory.
In addition, I have edited the Windows System variable PATH and added in the new \BIN\ directory;
I have also downloaded the Visual C++ Redistributable Packages for Visual Studio 2013, the 32 bit version I found from here:
https://www.microsoft.com/en-us/download/details.aspx?id=40784
When I execute I first get a library not found. Then on subsequent executions I get the error "Execute failed: Could not initialize class com.tableausoftware.extract.ExtractAPI"
Does anyone know how to properly set up KNIME's environment to successfully use the Tableau connector?
Okay here are the steps I used in order.
Download the Java version of the Tableau SDK for Windows:
https://onlinehelp.tableau.com/current/api/sdk/en-us/help.htm#SDK/tableau_sdk_installing.htm%3FTocPath%3D_____3
Unzip that file into a directory called SDK in the workspace you are using. For me this was C:\Users\<'user name'>\knime-workspace\
Update environment variables for "Path" on your system and in KNIME add a linked folder resource (Preferences -> Workspace -> Linked Resources)
Point both towards the \bin\ sub-directory.
THEN installed the "Microsoft Visual Basic 2013 Redistributables" file.
I'm using Weka 3.6.11 and I want to use its LibLinear wrapper.
I get the message that "Liblinear classes not in CLASSPATH"
I am on Windows. I create the CLASSPATH to the system variables, and wrote the path to the liblinear.jar file which happens to be
C:\Program Files (x86)\Weka-3-6\LibLINEAR.jar
So now, I'm not sure what the problem is. Any ideas?
I haven't exactly resolved my issue but when I used Weka 3.7 instead of 3.6 and used the Weka Package manager to install LibLinear and LIBSVM, they worked just fine.
I download liblinear.1.92 from http://liblinear.bwaldvogel.de/ , put the .jar file under
$HOME/lib/liblinear-1.92/liblinear-1.92.jar
and start weka as
java -classpath $CLASSPATH:weka.jar:$HOME/lib/libsvm-3.16/java/libsvm.jar:$HOME/lib/liblinear-1.92/liblinear-1.92.jar weka.gui.GUIChooser
every thing was fine until I wanted to use liblinear classifier, weka shows "liblinear not in CLASSPATH". What could be the problem? Thanks.
That's because the liblinear.jar you downloaded is not the original one. The creator of that jar uses his own package structure, his classes are placed in
de.bwaldvoge.liblinear
but the standard weka classifier looks for Linear in
liblinear
On the bottom of bwaldvogels website that you linked, he gives a link to a Weka wrapper that points to his package.
The Weka package manager is available from version 3.7.2 or later, so you have to download the developer version, since the latest stable is 3.6.9.
Then, simply start the Weka package manager and install LibLINEAR. It worked for me this way. Note, that this will create a "wekafiles" directory in $HOME.