Develop Reference

Ensuring files are available to the JVM

I'm trying to install TensorFlow for Java on Windows 10 using this Article
. I followed the steps carefully but the windows commands didn't work with me so I decided to do it manually.
The first command is to make the .jar part of the classpath and I did it manually
but the second step was to ensure that the following two files are available to the JVM: the .jar file and the extracted JNI library
but I don't know how to do that manually
The code:
package securityapplication;
import org.tensorflow.TensorFlow;
import org.tensorflow.Graph;
import org.tensorflow.Session;
import org.tensorflow.Tensor;
public class SecurityApplication {
public static void main(String[] args) throws Exception {
try (Graph g = new Graph()) {
final String value = "Hello from " + TensorFlow.version();
// Construct the computation graph with a single operation, a constant
// named "MyConst" with a value "value".
try (Tensor t = Tensor.create(value.getBytes("UTF-8"))) {
// The Java API doesn't yet include convenience functions for adding operations.
g.opBuilder("Const", "MyConst").setAttr("dtype", t.dataType()).setAttr("value", t).build();
}
// Execute the "MyConst" operation in a Session.
try (Session s = new Session(g);
Tensor output = s.runner().fetch("MyConst").run().get(0)) {
System.out.println(new String(output.bytesValue(), "UTF-8"));
}
}
}
}
could someone help? cuz my program that uses TensorFlow still have the following error
The text in the image is :
Exception in thread "main" java.lang.UnsatisfiedLinkError: Cannot find TensorFlow native library for OS: windows, architecture: x86. See https://github.com/tensorflow/tensorflow/tree/master/tensorflow/java/README.md for possible solutions (such as building the library from source). Additional information on attempts to find the native library can be obtained by adding org.tensorflow.NativeLibrary.DEBUG=1 to the system properties of the JVM.
at org.tensorflow.NativeLibrary.load(NativeLibrary.java:66)
at org.tensorflow.NativeLibrary.load(NativeLibrary.java:66)
at org.tensorflow.TensorFlow.init(TensorFlow.java:36)
at org.tensorflow.TensorFlow.<clinit>(TensorFlow.java:40)
at org.tensorflow.Graph.<clinit>(Graph.java:194)
at securityapplication.SecurityApplication.main(SecurityApplication.java:15) Java Result: 1 BUILD SUCCESSFUL (total time: 4 seconds)
The result after running the first command in cmd:
The result after running the second command in Windows PowerShell:
Any suggestions?!
Thank you

The first command failure (javac) suggests that the javac command is not in your PATH environment variables. See for example, this StackOverflow question
For the second command failure, I believe the space after -D is what is causing you trouble as Holger suggested.
IDEs like Eclipse and others also provide a means to set the java.library.path property for the JVM (see this StackOverflow answer for example).
Background: TensorFlow for Java consists of a Java library (packaged in a .jar file) and a native library (.dll on Windows, distributed in a .zip file). You need to ensure that the .jar file is in the classpath and the directory containing the .dll is in included in the java.library.path of the JVM when executing a program.
Hope that helps.

Unable to Load Dependent SO file in LInux

I am new to linux. I am trying to load a SO file in Ubuntu using Java. The file that I have specified in the java method "System.load(/home/ab/Downloads/libtesseract.so)" loads fine but its dependent so file placed in the same place as "libtesseract.so" is not found. Here is the error message I get. Error: UnSatisfiedLinkError and says "liblept.so.4" cound not be found. This so file is placed in the same location as libtesseract.so. When I place "liblept.so.4" in the "/lib". It is able to load this so file from. So what I understood is that for, its not for java to load the dependent so. It has to be loaded by ubuntu. So I tried a simple application to load this by setting the PATH variable with the location of the so file. And exported the java code into a jar and tried to run this jar file from terminal as the path variable is not persistent for entire system. It worked fine. So I tried to do the same thing programmatically by using the code below to its not working. Please advice. TIA
Code:
ProcessBuilder pb = new ProcessBuilder("/bin/sh");
Map<String, String> envMap = pb.environment();
envMap.put("LD_LIBRARY_PATH", "/home/ab/Downloads");
envMap.put("PATH", "/home/ab/Downloads");
Set<String> keys = envMap.keySet();
for(String key:keys)
{
System.out.println(key+" ==> "+envMap.get(key));
}
System.load("/home/ab/Downloads/libtesseract.so");

As far as I know you can't really modify the environment variables in Java "on-the-fly". That means you should set both LD_LIBRARY_PATH and PATH before running the java.

unable to execute hadoop fs -put command from Java

I am trying to execute hadoop fs -put <source> <destination> from Java code. When I execute this command directly from the terminal, it works fine but when I try to execute this command from within the Java code using
String[] str = {"/usr/bin/hadoop","fs -put", source, dest};
Runtime.getRuntime().exec(str);
I get error as Error: Could not find or load main class fs. I tried to execute some non-hadoop commands like ls,mkdir commands from Java and they worked fine but the hadoop commands are not getting executed even though they work fine from the terminal.
What could be the possible reason for this and how can I solve it?
JAVA API TRY: I tried to use java api to perform the copy operation but I get error. The Java code is :
String source = "/home/tmpe/file1.csv";
String dest = "/user/tmpe/file1.csv";
Configuration conf = new Configuration();
conf.set("fs.defaultFS", "hdfs://node1:8020");
FileSystem fs = FileSystem.get(conf);
Path targetPath = new Path(dest);
Path sourcePath = new Path(source);
fs.copyFromLocalFile(false,true,sourcePath,targetPath);
The error which I get is:
Exception in thread "main" java.io.IOException: Mkdirs failed to create /user/tmpe
at org.apache.hadoop.fs.ChecksumFileSystem.create(ChecksumFileSystem.java:378)
at org.apache.hadoop.fs.ChecksumFileSystem.create(ChecksumFileSystem.java:364)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:564)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:545)
at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:452)
at org.apache.hadoop.fs.FileUtil.copy(FileUtil.java:229)
at org.apache.hadoop.fs.FileSystem.copyFromLocalFile(FileSystem.java:1230)
I have already created /user/tmpe folder and it has full read-write permissions but still this error comes. I am unable to get the issue resolved

I guess you probably do not have a HADOOP_HOME environment variable set.
But since you're in Java, why on earth would you want to do a haddop fs -put in an external process when the Java API is even more friendly than the shell ?

Came across old post but if you haven't tried already, execute it with hadoop jar app_name.jar instead of java -jar. This way if classpath of your jar does not have all hadoop jars it will fetch the jars predefined in $HADOOP_CLASSPATH.

Run JAR file in PHP and write a file

I have the following problem in php when excecuting a jar file. I use the following command:
exec("java -jar JavaProject4.jar";
The JavaProject4.jar creates a txt file in a path given in the java code.
When i run the project in NetBeans the txt file is created. However, when i excecute the jar in php i don't get any errors but i can't get the file.
Here the java code I use to write the file:
public static void main(String[] args) throws InterruptedException, FileNotFoundException, IOException {
Main a = new Main();
List<Double> l1 = new ArrayList<Double>();
l1 = a.compute_features();
//System.out.println(l1);
FileWriter fstream1 = new FileWriter("C:/wamp/www/test/out.txt");
BufferedWriter out1 = new BufferedWriter(fstream1);
out1.write(l1.toString());
out1.newLine();
out1.close();
}
Im using a wamp server with php 5.2.4 and the latest java version.
Thanks a lot!
EDIT:
Problem solved, I moved the main java file in NetBeans to the default package and also fixed a wrong path and now everything is working as expected.
Thanks everyone

When you run it with PHP, how are you doing so? Are you using the PHP CLI (Command Line Interface), or are you running it through an Apache Module (CGI or otherwise)? The reason I ask, is because the problem you are having could have something to do with the user who the script is executing as. If you are using the CLI, you are running as your Windows User, however, if you are running it through Apache, then it is running as whatever user Apache is running as. Therefore, you might need to give the relevant permissions to the Apache user for that directory you are writing to.
Regards,
Ralfe

I think that you need to specify the full path for :
- the java executable
- your jar
Example :
exec("/usr/bin/java -jar /my/java/project/path/JavaProject4.jar");

Running java code on VMWare with references to external jar files

I am trying to run simple java code on VMWare Workstation. I have the following simple test Main file:
import cern.jet.random.engine.RandomSeedGenerator;;
public class TestDataService {
//private static Logger logger = Logger.getLogger(TestDataService.class);
/**
* #param args
*/
public static void main(String[] args) {
// TODO Auto-generated method stub
System.out.println("Hello World DAI!");
// Input some data.
RandomSeedGenerator re = new RandomSeedGenerator();
return;
}
}
RandomSeedGenerator is a class in colt.jar library, and I have the jar file under my lib folder.
I am building the project with ant, and I have the following manifest file where I set the classpath:
Manifest-Version: 1.0
Main-Class: edu.umass.TestDataService
Name: edu/umass/TestDataService/Version.class
Class-Path: lib/colt.jar
When I run the code from the VMWare shell which runs Red Hat Linux, I get this Exception:
[root#localhost] java -jar app.jar
Hello World DAI!
Exception in thread "main" java.lang.NoClassDefFoundError: cern/jet/random/engine/RandomSeedGenerator
at edu.umass.TestDataService.main (Unknown Source)
Caused by: java.long.ClassNotFoundException: cern.jet.random.engine.RandomSeedGenerator
Just as a final note, everything seems to work fine on windows with eclipse, but nothing seems to work on the virtual machine. Any ideas?

Did you install the jar files required by your application on the VMs?
Did you configured CLASS_PATH correctly?

I doubt there is an issue with the jvm or the vm. The problem is going to be in how you run the class. Specifically how your setting the classpath. Try this:
Navigate to where you've placed colt.jar. Get the present working directory by typing in pwd. Use this to construct the run command using the absolute path to colt.jar.
So eventually you should be running (from the directory containing your jar) something like this:
java -cp /the/full/path/to/lib/colt.jar -jar app.jar
Once you've got that work you can then try and figure out what the correct relative path is. and then you'll be able to do
java -cp a/relativel/path/to/lib/colt.jar -jar app.jar