I created my weka model in the machine and imported it to the android project. When i try to create the classifier it gives an error "exception.java.io.StreamCorruptedException" when i try to deserialise the model i created. The code perfectly works in machine.
This is my Code,
InputStream fis = null;
fis = new InputStream("/modle.model");
InputStream is = fis;
Classifier cls = null;
//here im getting the error when trying to read the Classifier
cls = (Classifier) SerializationHelper.read(is);
FileInputStream datais = null;
datais = new FileInputStream("/storage/emulated/0/window.arff");
InputStream dataIns = datais;
DataSource source = new DataSource(dataIns);
Instances data = null;
try {
data = source.getDataSet();
} catch (Exception e) {
e.printStackTrace();
}
data.setClassIndex(data.numAttributes() - 1);
Instance in = new Instance(13);
in.setDataset(data);
in.setValue(0, testWekaModle1[0]);
in.setValue(1, testWekaModle1[1]);
in.setValue(2, testWekaModle1[2]);
in.setValue(3, testWekaModle1[3]);
in.setValue(4, testWekaModle1[4]);
in.setValue(5, testWekaModle1[5]);
in.setValue(6, testWekaModle1[6]);
in.setValue(7, testWekaModle1[7]);
in.setValue(8, testWekaModle1[8]);
in.setValue(9, testWekaModle1[9]);
in.setValue(10, testWekaModle1[10]);
in.setValue(11, testWekaModle1[11]);
double value = 0;
value = cls.classifyInstance(in);
in.setClassValue(value);
This is the full stacktrace,
java.io.ObjectInputStream.readStreamHeader(ObjectInputStream.java:2109)
java.io.ObjectInputStream.<init>(ObjectInputStream.java:372)
weka.core.SerializationHelper.read(SerializationHelper.java:288)
info.androidhive.sleepApp.model.ControllerWeka.wekaModle(ControllerWeka.java:81)
info.androidhive.sleepApp.activity.HomeFragment.extract(HomeFragment.java:278)
info.androidhive.sleepApp.activity.HomeFragment.stop(HomeFragment.java:146)
"info.androidhive.sleepApp.activity.HomeFragment$2.onClick(HomeFragment.java:107)"
android.view.View.performClick(View.java:4475)"
android.view.View$PerformClick.run(View.java:18786)"
android.os.Handler.handleCallback(Handler.java:730)"
dalvik.system.NativeStart.main(Native Method)"
com.android.internal.os.ZygoteInit.main(ZygoteInit.java:1025)"
com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:1209)"
java.lang.reflect.Method.invoke(Method.java:525)"
java.lang.reflect.Method.invokeNative(Native Method)"
android.app.ActivityThread.main(ActivityThread.java:5419)"
android.os.Looper.loop(Looper.java:137)"
android.os.Handler.dispatchMessage(Handler.java:92)"
Please help me to overcome this problem.
this is resolved, the model was created in a different environment(PC) and tried to deserialise in the android environment which gave error because of the two types of JDK wasn't same at all.
Be sure that both of the weka.jar have the same version.
And do NOT use the GUI version of Weka to save the model since the Android runtime does not contain GUI related packages used by weka.
It would be fine that build and save the model programmatically with desktop and deserialise it through Android.
Related
I am trying to read the contents of a PDF file using Java-Selenium. Below is my code. getWebDriver is a custom method in the framework. It returns the webdriver.
URL urlOfPdf = new URL(this.getWebDriver().getCurrentUrl());
BufferedInputStream fileToParse = new BufferedInputStream(urlOfPdf.openStream());
PDFParser parser = new PDFParser((RandomAccessRead) fileToParse);
parser.parse();
String output = new PDFTextStripper().getText(parser.getPDDocument());
The second line of the code gives compile time error if I don't parse it to RandomAccessRead type.
And when I parse it, I get this run time error:
java.lang.ClassCastException: java.io.BufferedInputStream cannot be cast to org.apache.pdfbox.io.RandomAccessRead
I need help with getting rid of these errors.
First of, unless you want to interfere in the PDF loading process, there is no need to explicitly use the PdfParser class. You can instead use a static PDDocument.load method:
URL urlOfPdf = new URL(this.getWebDriver().getCurrentUrl());
BufferedInputStream fileToParse = new BufferedInputStream(urlOfPdf.openStream());
PDDocument document = PDDocument.load(fileToParse);
String output = new PDFTextStripper().getText(document);
Otherwise, if you do want to interfere in the loading process, you have to create a RandomAccessRead instance for your BufferedInputStream, you cannot simply cast it because the classes are not related.
You can do that like this
URL urlOfPdf = new URL(this.getWebDriver().getCurrentUrl());
BufferedInputStream fileToParse = new BufferedInputStream(urlOfPdf.openStream());
MemoryUsageSetting memUsageSetting = MemoryUsageSetting.setupMainMemoryOnly();
ScratchFile scratchFile = new ScratchFile(memUsageSetting);
PDFParser parser;
try
{
RandomAccessRead source = scratchFile.createBuffer(fileToParse);
parser = new PDFParser(source);
parser.parse();
}
catch (IOException ioe)
{
IOUtils.closeQuietly(scratchFile);
throw ioe;
}
String output = new PDFTextStripper().getText(parser.getPDDocument());
(This essentially is copied and pasted from the source of PDDocument.load.)
I'm trying to use Apache OpenNLP POSTagger example codes, and i've come up with an error, and below is the code
public String[] SentenceDetect(String qwe) throws IOException
{
POSModel model = new POSModelLoader().load(new File("/home/jebard/chabacano/Chabacano1/src/en-pos-maxent.bin"));
PerformanceMonitor perfMon = new PerformanceMonitor(System.err, "sent");
POSTaggerME tagger = new POSTaggerME(model);
String input = "Hi. How are you? This is Mike.";
ObjectStream<String> lineStream = new PlainTextByLineStream(
new StringReader(input));
perfMon.start();
String line;
while ((line = lineStream.read()) != null) {
String whitespaceTokenizerLine[] = WhitespaceTokenizer.INSTANCE
.tokenize(line);
String[] tags = tagger.tag(whitespaceTokenizerLine);
POSSample sample = new POSSample(whitespaceTokenizerLine, tags);
System.out.println(sample.toString());
perfMon.incrementCounter();
}
perfMon.stopAndPrintFinalResult();
Error at this line
.load(new File("/home/jebard/chabacano/Chabacano1/src/en-pos-maxent.bin")
The method load(java.io.File) in type ModelLoader is not applicable for the arguments(org.apache.tomcat.jni.File)
This is actually not a bug in OpenNLP. It's a bug in your code, as you load the class File from the package (aka namespace) org.apache.tomcat.jni.File.
Yet, the API of OpenNLP requests you to use the class File from the standard JDK package java.io, i.e. you should import java.io.File instead.
In general, this should fix your problem.
Important hint
You should migrate your code, as models should not be loaded via POSModelLoader
Loads a POS Tagger Model for the command line tools.
Note: Do not use this class, internal use only!
Instead you can use the constructor POSModel(InputStream in) to load your model file via an InputStream referencing the actual model file.
Moreover, the class POSModelLoader was only present in previous releases of OpenNLP (versions <= 1.5.x). In the latest OpenNLP version 1.6.0 it was removed completely. Instead you can and should now use the constructor of the POSModel class to load/initialize the model you need.
There is some problem with XML parsing. Try this, it worked for me.
System.setProperty("org.xml.sax.driver", "org.xmlpull.v1.sax2.Driver");
try {
AssetFileDescriptor fileDescriptor =
context.getAssets().openFd("en_pos_maxent.bin");
FileInputStream inputStream = fileDescriptor.createInputStream();
POSModel posModel = new POSModel(inputStream);
posTaggerME = new POSTaggerME(posModel);
} catch (Exception e) {}
This is a libgdx project using gradle if that matters. I have a class that implements screen where I want to draw a string to the display. on Desktop the following code works.
try {
string = readFile("bin/001.txt",Charset.defaultCharset());
} catch (IOException e) {
e.printStackTrace();
}
And then I can use that string in a Label for a stage. However on the android version I use "assets/001.txt" and get the error VFY: unable to resove static method 11542: Ljava/nio/file/Paths;.g
I then researched it and found a suggestion here
FileInputStream fis;
fis = openFileInput("test.txt");
StringBuffer fileContent = new StringBuffer("");
byte[] buffer = new byte[1024];
while ((n = fis.read(buffer)) != -1)
{
fileContent.append(new String(buffer, 0, n));
}
This did not work either because openFileInput is undefined because "Those are methods defined on the Context class" however
"If you are using LibGDX, you need to forego the concept of Activities and Views, since your entire game will now just be a single Activity. To have a main menu when you game starts up doesn't switch the Activity, but just presents a different set of objects to be rendered to the same Activity."
And now I'm lost
To read a file in LibGdx you need to abstract from the underlying. This is because the same application will work in several systems (Desktop, Android, Web, ...).
So to read a file you need to use:
FileHandle file = Gdx.files.internal("myfile.txt");
String text = file.readString();
The path and location of the files can be summarized in the documentation
I am trying to work with Open Office in my Java app.
Based on SDK, with extra help from bootstrapconnector.jar I successfully started empty swritter and can write to the document.
Now, I would like to open document stored in ByteArray and after some modifications save changes doc to ByteArray.
Can somebody help me doing that, please?
Here is the SDK part starting sWritter.
public static com.sun.star.text.XTextDocument openWriter(
com.sun.star.uno.XComponentContext xContext) {
//define variables
com.sun.star.frame.XComponentLoader xCLoader;
com.sun.star.text.XTextDocument xDoc = null;
com.sun.star.lang.XComponent xComp = null;
try {
// get the remote office service manager
com.sun.star.lang.XMultiComponentFactory xMCF =
xContext.getServiceManager();
Object oDesktop = xMCF.createInstanceWithContext(
"com.sun.star.frame.Desktop", xContext);
xCLoader = (com.sun.star.frame.XComponentLoader) UnoRuntime.queryInterface(com.sun.star.frame.XComponentLoader.class,
oDesktop);
com.sun.star.beans.PropertyValue[] szEmptyArgs =
new com.sun.star.beans.PropertyValue[0];
String strDoc = "private:factory/swriter";
xComp = xCLoader.loadComponentFromURL(strDoc, "_blank", 0, szEmptyArgs);
xDoc = (com.sun.star.text.XTextDocument) UnoRuntime.queryInterface(com.sun.star.text.XTextDocument.class,
xComp);
} catch (Exception e) {
System.err.println(" Exception " + e);
e.printStackTrace(System.err);
}
return xDoc;
}
as you can see there is a method loadComponentFromURL.
I saw somewhere else, in the OOoBeanViewer, that it is possible to read and write doc to ByteArray, however I don't know how to achieve that without officebean.jar which I don't want to use in my project.
Thanks for your comments and hints.
While trying to find an answer to Android Jasper Reporting I found out that there are two other questions to be answered therefor, which I been asked to ask as a question, not as an answer ;):
My questions are now: "Is there any compiler to use directly on the device" AND "how to execute such without rooting the device.
If anybody could give me a hint I would really appreciate it...
I looked a little time forward on this approach, and found apps which makes it possible to create APKs directly on an Android device which is NOT rooted:
TerminalIDE - https://play.google.com/store/apps/details?id=com.spartacusrex.spartacuside&hl=de
JavaIDEdroid - http://code.google.com/p/java-ide-droid/
AIDE - https://play.google.com/store/apps/details?id=com.aide.ui&hl=en
Looks like they're using the compiler from eclipse and a ported dex converter. Now I'm trying to figure out how to do the same.
Sure: get the source code and look into it. But while I'm having curious problems to get a connection to the servers and trying to solve it, I follow the plea to ask this question here. Hoping both to help others with it and also getting an answer for myself ;)
I took the org.eclipse.jdt.core_3.7.3.v20120119-1537.jar from the plugin directory of my indigo and tried following code:
org.eclipse.jdt.internal.compiler.batch.Main ecjMain = new org.eclipse.jdt.internal.compiler.batch.Main(new PrintWriter(System.out), new PrintWriter(System.err), false/*noSystemExit*/, null, progress);
System.err.println("compiling...");
ecjMain.compile(new String[] {"-classpath", "/system/framework", storage.getAbsolutePath()+"/Test.java"});
ecjMain.compile(new String[] {storage.getAbsolutePath()+"/Test.java"});
System.err.println("compile succeeded!!!");
Sometimes the Exception was thrown that java.lang.Object could not be found and othertimes it stuck doing nothing while heating up my processor with 100% usage ... ...
At this time i could not figure out what is happening and why. And in cause that i have other work to do this part has to wait a little.
I succeeded after taking inspiration from source of JavaIDEdroid and realizing that I'm dumb (for a time I tried to uses the compiler with the dexified framework classes on the device - which naturtally could not work).
After i succeeded compiling my Test.java with a copy of ADTs android-jar on sdcard I just had to load the classes with the DexClassLoader.
While informing myselft about how to do that I found this nice article Custom Class Loading in Dalvik which inspired me at least to write this piece of code:
File storage = getDir("all41", Context.MODE_PRIVATE);
System.err.println("copying the android.jar from asssets to the internal storage to make it available to the compiler");
BufferedInputStream bis = null;
OutputStream dexWriter = null;
int BUF_SIZE = 8 * 1024;
try {
bis = new BufferedInputStream(getAssets().open("android.jar"));
dexWriter = new BufferedOutputStream(
new FileOutputStream(storage.getAbsolutePath() + "/android.jar"));
byte[] buf = new byte[BUF_SIZE];
int len;
while((len = bis.read(buf, 0, BUF_SIZE)) > 0) {
dexWriter.write(buf, 0, len);
}
dexWriter.close();
bis.close();
} catch (Exception e) {
System.err.println("Error while copying from assets: " + e.getMessage());
e.printStackTrace();
}
System.err.println("instantiating the compiler and compiling the java file");
org.eclipse.jdt.internal.compiler.batch.Main ecjMain = new org.eclipse.jdt.internal.compiler.batch.Main(new PrintWriter(System.out), new PrintWriter(System.err), false/*noSystemExit*/, null);
ecjMain.compile(new String[] {"-classpath", storage.getAbsolutePath()+"/android.jar", Environment.getExternalStorageDirectory().getAbsolutePath() + "/Test.java"});
System.err.println("calling DEX and dexifying the test class");
com.android.dx.command.Main.main(new String[] {"--dex", "--output=" + storage.getAbsolutePath() + "/Test.zip", Environment.getExternalStorageDirectory().getAbsolutePath() + "/./Test.class"});
System.err.println("instantiating DexClassLoader, loading class and invoking toString()");
DexClassLoader cl = new DexClassLoader(storage.getAbsolutePath() + "/Test.zip", storage.getAbsolutePath(), null, getClassLoader());
try {
Class libProviderClazz = cl.loadClass("Test");
Object instance = libProviderClazz.newInstance();
System.err.println(instance.toString());
} catch (Exception e) {
System.err.println("Error while instanciating object: " + e.getMessage());
e.printStackTrace();
}
The Test.java only contains one method:
public String toString() {
return "Hallo Welt!";
}
To get it running you need the jars jdt-compiler-x.x.x.jar (found in plugins directory of eclipse) and dx.jar (found in directory platform-tools/lib of Android SDK)
Not really hard ;) And now I will find out what to change in source of JasperReports to get it work on our beloved Android devices :D