Bob Matthews

Hi Tim

I have just realised that your comments re Maven Central apply to deeplearning4j NOT wekadeeplearning4j which is the package I am trying to access from my java program

Bob M

error output:-

FAILURE: Build failed with an exception.

*What went wrong:
java.io.IOException: The filename, directory name, or volume label syntax is incorrect

--scan option seems to be the fullest output

yes

gradlew build -x test PublishToMavenLocal

is where my build falls over with errors

I have added --debug, --stacktrace. --info options but the output is overwhelming
Am wondering whether there are other prerequisites that I need to do ?

Bob M

p.s. have asked a question on  a Gradle forum also - no replies as yet

Hi Knute and Tim

Thanks for persevering

The person I am talking to at weka is not the author of this package and runs Linux - so no help

When I first ran the 3rd line gradlew build etc, it seemed to perform most of the tasks and then fell over with error messages which I can't reproduce

when I rerun that line I am now getting an IOException error

as usual, I know nothing about using gradle but I do need to get this package onto my machine somewhere, somehow

Bob M

p.s. I am 73 yoa - thus the strange questions from time to time

Hi Knute

yes - When I did try, I tried as you suggested but got many errors

I have gone back to the weka people with some questions

Bob M

Thanks Tim

I tried your suggestion but same error

Cheers

Bob M

I have tried the below code but the 3rd line failed

It is also possible to include this package as maven project.
As of now it is not provided in any maven repository,
therefore you need to install this package to your local .m2 repository:

$ git clone https://github.com/Waikato/wekaDeeplearning4j.git $ cd wekaDeeplearning4j $ ./gradlew build -x test publishToMavenLocal

I understand I need to add a repository for the weka dependency but do not know which repository to add,
and I don't understand how the weka dependency knows to utilize the missing weka repository

Bob M

Thank you Knute for your recent advice

I am slowly getting there

I have opened up a project in Netbeans which was created a year or more ago called JForex-3-SDK and then opened the pom.xml file.

It starts with a comment - it is an example maven pom file which allows the user to build JForex-SDK by using the latest version of dependencies from Dukascopy public repository with command: mvn clean install

Two of the dependencies viz. com.dukascopy.dds2 and com.dukascopy.api have versions which I have updated

I also added the dependency you advised wekaDeeplearning4j version 1.5.14

I then ran "Clean and Build Project"

Output is as follows:-

Failed to execute goal on JForex-3-SDK: Could not resolve dependencies for project com.dukascopy.dds2:JForex-3-SDK.jar:3.6.14: Failure to find nz.ac.waikato.cms.weka.Deeplearning4j:jar:1.5.134 in http://www.dukascopy.com/client/jforexlib/publicrepo/ was cashed in the local repository.....etc.

I can see that it is looking in the dukascopy repository for the Deeplearning4j dependency as that is the only repository in the pom.xml file

I don't know what the trouble is with the dds2 dependency ??

Below is a portion of the rather long pom.xml

<?xml version="1.0" encoding="UTF-8"?> <!-- The following is an example Maven pom file which allows the user to to build JForex-SDK by using the latest version of dependencies from dukascopy public repository with command: mvn clean install --> <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd"> <modelVersion>4.0.0</modelVersion> <groupId>com.dukascopy.dds2</groupId> <artifactId>JForex-3-SDK</artifactId>    <version>3.6.14</version> <organization> <name>Dukascopy Bank SA</name> <url>http://www.dukascopy.com</url> </organization> <repositories> <repository> <id>dc_public</id> <name>Dukascopy public repository</name> <url>http://www.dukascopy.com/client/jforexlib/publicrepo/</url> </repository> </repositories> <dependencies> <dependency> <groupId>com.dukascopy.dds2</groupId> <artifactId>DDS2-jClient-JForex</artifactId> <!-- find the latest available version here: http://www.dukascopy.com/client/jforexlib/publicrepo/com/dukascopy/dds2/DDS2-jClient-JForex/ --> <version>3.6.14</version> </dependency> <dependency> <groupId>com.dukascopy.api</groupId> <artifactId>JForex-API</artifactId> <!-- find the latest available version here:                http://www.dukascopy.com/client/jforexlib/publicrepo/com/dukascopy/api/JForex-API/            --> <version>2.13.88</version> <classifier>sources</classifier> </dependency>                <dependency>                    <groupId>nz.ac.waikato.cms.weka</groupId>                    <artifactId>wekaDeeplearning4j</artifactId>                          <version>1.5.14</version>                </dependency>        </dependencies>

Thanks Knute

I have not used Maven before........................

The weka people told me that...........This Weka package itself [wekadeeplearning4j] is not a Maven project.
You will have to include all the jars in your project that are available through the release zip files.

So I was a little confused

Bob M

p.s. I have the following out-of-date instructions to follow:-

Using wekaDeeplearning4j in a Maven Project

It is also possible to include this package as maven project.
As of now it is not provided in any maven repository - this is INCORRECT NOW,
therefore you need to install this package to your local .m2 repository:

$ git clone https://github.com/Waikato/wekaDeeplearning4j.git $ cd wekaDeeplearning4j $ ./gradlew build -x test publishToMavenLocal

Now you can add the maven dependency in your pom.xml file

<dependency>    <groupId>nz.ac.waikato.cms.weka</groupId>    <artifactId>wekaDeeplearning4j</artifactId>    <version>${wekaDeeplearning4j.version}</version> </dependency>

so - question - what instructions should I follow ?

Update: have installed Maven and on reading it seems that I should create a new project (which I have done in Netbeans)
but am now told to add dependencies by updating the pom.xml file - and here I am lost

Hi Paul

My limited understanding is as follows.....................

Of the many classifiers available in weka, this one - Deeplearning4jMlpClassifier is different in that the code for it is a third party open source

The weka people have created a wrapper for this classifier (whatever that means) but the one main weka.jar file I have used in the past for other classifiers does NOT include the many .jar dependencies for this classifier

And so I have to bring them in myself manually (I think)

Reading the Dukascopy page seems to suggest that I might have to include all 169 .jar files (less those I can identify are for MAC machines, LINUX etc) one by one - a long arduous process

Thus my original question

Bob M

Hi Paul

Have tried but without success................

The following is from Dukascopy's website

Include libraries

In order to use an external library one has to add a @Library annotation to the strategy. There are two options how to refer to the libaries:

   By name, for instance:

  @Library("myLib.jar")    public class Strategy implements IStrategy {

   in this case the library has to be placed in the strategy files' directory, i.e., IContext.getFilesDir().

   By full path, for instance:

   @RequiresFullAccess    @Library("C:\\temp\\myLib.jar")    public class Strategy implements IStrategy {

in this case one also has to add the @RequiresFullAccess annotation.
************************************************************
I use the latter method which requires FullAccess

when I try something like "C:/.........../wekaDeeplearning4j_v1.5.14/lib/*" and then compile on Dukascopy's platform it says it can't find that path ???

Bob M

Thank's Paul

Bob M

Hi

I have various java programs which utilize weka mining software out of Waikato University, New Zealand

My programs are compiled and run on Dukascopy's JForex platform

In order to utilize certain weka classifiers, I have to include a handful of .jar files in my programs

I am now experimenting with a new weka package for deeplearning4j - includes the dl4jMlpClassifier

This package is not a maven project and I have been advised that I will have to include all the jars in my project that are available through the release zip files.

When I look at the wekaDeeplearning_v1.5.14 lib folder I see 169 .jar files

How can I easily include such a large number of .jar files into my java program ?

Thank you

Bob M

New Zealand

hi

i had a class that was not invoked (not coded as yet)

within that class lay the code

Bob M

my bad

the code was within a class that was not called !

Bob M