How to get text paragraph dl4j neural networking - java

I'm working deeplearning4j and not understand how to get text paragraph of vector classified from the neural network.
I could only get classification rate.
This is my code:
public static void main(String[] args) throws Exception {
ClassPathResource resource = new ClassPathResource("paravec/recortes");
LabelAwareIterator iterator = new FileLabelAwareIterator.Builder()
.addSourceFolder(resource.getFile()).build();
TokenizerFactory t = new DefaultTokenizerFactory();
t.setTokenPreProcessor(new CommonPreprocessor());
ParagraphVectors paragraphVectors = new ParagraphVectors.Builder()
.learningRate(0.025).minLearningRate(0.001).batchSize(1000)
.epochs(10).iterate(iterator).trainWordVectors(true)
.tokenizerFactory(t).build();
paragraphVectors.fit();
ClassPathResource unlabeledResource = new ClassPathResource(
"paravec/caderno");
FileLabelAwareIterator unlabeledIterator = new FileLabelAwareIterator.Builder()
.addSourceFolder(unlabeledResource.getFile()).build();
MeansBuilder meansBuilder = new MeansBuilder(
(InMemoryLookupTable<VocabWord>) paragraphVectors
.getLookupTable(),
t);
LabelSeeker seeker = new LabelSeeker(iterator.getLabelsSource()
.getLabels(),
(InMemoryLookupTable<VocabWord>) paragraphVectors
.getLookupTable());
while (unlabeledIterator.hasNextDocument()) {
LabelledDocument document = unlabeledIterator.nextDocument();
//how to get text paragraph?
INDArray documentAsCentroid = meansBuilder
.documentAsVector(document);
}
}
Thanks!
Renan.

Related

How can I use custom font in Jeuclid (Java)?

this is the code where I am converting MathMl to png
public static void main(String[] args) throws IOException, FontFormatException {
Converter converter = Converter.getInstance();
String math="<math xmlns=\"http://www.w3.org/1998/Math/MathML\"><mfenced open=\"[\" close=\"]\"><mtable><mtr><mtd><mn>3</mn></mtd><mtd><mn>2</mn></mtd></mtr><mtr><mtd><mn>6</mn></mtd><mtd><mn>7</mn></mtd></mtr><mtr><mtd><mn>6</mn></mtd><mtd><mn>4</mn></mtd></mtr></mtable></mfenced><mo>=</mo><msubsup><mo>∫</mo><mn>5</mn><mn>4</mn></msubsup></math>";
File inputFile = new File("D:\\mathml.xml");
File outputFile = new File("D:\\image.jpg");
//params to mention the size of image
MutableLayoutContext params = new LayoutContextImpl(
LayoutContextImpl.getDefaultLayoutContext());
params.setParameter(Parameter.MATHSIZE, 50f);
Document doc = StringToDocumentToString.convertStringToDocument(math);
// Parameter parameter= new Pa
converter.convert(doc, outputFile , "image/jpeg", params);
}
I want to use Tahoma.ttf when I make png from MathMl but I can not find any resource how to do that. Please can anyone help me?

Delete the READ access on that topic with the principal using java code

public void deleteReadAcl(AdminClient adminClient, TopicReadRequest deleteAclRequest) {
ResourcePatternFilter resourcePatternFilter = new ResourcePatternFilter(ResourceType.GROUP, deleteAclRequest.getConsumerGroup(), PatternType.ANY);
AccessControlEntryFilter accessControlEntryFilter = new AccessControlEntryFilter(deleteAclRequest.getPrincipal(), null, AclOperation.READ,AclPermissionType.ALLOW);
AclBindingFilter aclBindingFilter = new AclBindingFilter(resourcePatternFilter, accessControlEntryFilter);
Set<AclBindingFilter> bindingFilters = new HashSet<>();
bindingFilters.add(aclBindingFilter);
System.out.println("acl to be delted === "+bindingFilters);
adminClient.deleteAcls(bindingFilters);

Adding hyperlink to inner PDF files

I have to create a PDF file by adding two PDF files inside a generated PDF file as a tree structure using iText in Java.
I have to create bookmarks with PDF file names and add a hyperlink to the bookmark. When the bookmark is clicked, the respective PDF should be opened in that PDF file itself, not as a separate PDF.
PDFTREE
pdf1
pdf2
Such bookmarks are referred to as outline elements in the PDF specification (PDF 32000-1:2008, p.367):
The outline consists of a tree-structured hierarchy of outline items (sometimes called bookmarks), which serve as a visual table of contents to display the document’s structure to the user.
If you merge the documents with PdfMerger, the outlines are copied to the resulting PDF by default. However, you want a main-node per document and not a flat list of bookmarks. Since cloning and copying outlines in no trivial task, it is best to let iText handle this. Unfortunately, we have little direct control how outlines are being merged.
We can build a SpecialMerger as a wrapper around PdfMerger to extract the cloned outlines (first step) and get them into a hierarchical structure afterwards (second step). The outline of each merged PDF is temporarily stored in the outlineList together with the desired name of the main node and its reference (page number in the merged PDF). After all the PDFs are merged, we can attach the temporarily stored outlines back to the root-node.
public static class SpecialMerger {
private final PdfDocument outputPdf;
private final PdfMerger merger;
private final PdfOutline rootOutline;
private final List<DocumentOutline> outlineList = new ArrayList<>();
private int nextPageNr = 1;
public SpecialMerger(final PdfDocument outputPdf) {
if (outputPdf.getNumberOfPages() != 0) {
throw new IllegalArgumentException("PDF must be empty");
}
this.outputPdf = outputPdf;
this.merger = new PdfMerger(outputPdf, true, true);
this.rootOutline = outputPdf.getOutlines(false);
}
public void merge(PdfDocument from, int fromPage, int toPage, String filename) {
merger.merge(from, fromPage, toPage); // merge with normal PdfMerger
// extract and clone outline of merged document
final List<PdfOutline> children = new ArrayList<>(rootOutline.getAllChildren());
rootOutline.getAllChildren().clear(); // clear root outline
outlineList.add(new DocumentOutline(filename, nextPageNr, children));
nextPageNr = outputPdf.getNumberOfPages() + 1; // update next page number
}
public void writeOutline() {
outlineList.forEach(o -> {
final PdfOutline outline = rootOutline.addOutline(o.getName()); // bookmark with PDF name
outline.addDestination(PdfExplicitDestination.createFit(outputPdf.getPage(o.getPageNr())));
outline.setStyle(PdfOutline.FLAG_BOLD);
o.getChildern().forEach(outline::addOutline); // add all extracted child bookmarks
});
}
private static class DocumentOutline {
private final String name;
private final int pageNr;
private final List<PdfOutline> childern;
public DocumentOutline(final String pdfName, final int pageNr, final List<PdfOutline> childern) {
this.name = pdfName;
this.pageNr = pageNr;
this.childern = childern;
}
public String getName() {
return name;
}
public int getPageNr() {
return pageNr;
}
public List<PdfOutline> getChildern() {
return childern;
}
}
}
Now, we can use this custom merger to merge the PDFs and then add the outline with writeOutline:
public static void main(String[] args) throws IOException {
String filename1 = "pdf1.pdf";
String filename2 = "pdf2.pdf";
try (
PdfDocument generatedPdf = new PdfDocument(new PdfWriter("output.pdf"));
PdfDocument pdfDocument1 = new PdfDocument(new PdfReader(filename1));
PdfDocument pdfDocument2 = new PdfDocument(new PdfReader(filename2))
) {
final SpecialMerger merger = new SpecialMerger(generatedPdf);
merger.merge(pdfDocument1, 1, pdfDocument1.getNumberOfPages(), filename1);
merger.merge(pdfDocument2, 1, pdfDocument2.getNumberOfPages(), filename2);
merger.writeOutline();
}
}
The result looks like this (Preview and Adobe Acrobat Reader on macOS):
Another option is to make a portfolio by embedding the PDFs. However, this is not supported by all PDF viewers and most users are not accustomed to these portfolios.
public static void main(String[] args) throws IOException {
String filename1 = "pdf1.pdf";
String filename2 = "pdf2.pdf";
try (PdfDocument generatedPdf = new PdfDocument(new PdfWriter("portfolio.pdf"))) {
Document doc = new Document(generatedPdf);
doc.add(new Paragraph("This PDF contains embedded documents."));
doc.add(new Paragraph("Use a compatible PDF viewer if you cannot see them."));
PdfCollection collection = new PdfCollection();
collection.setView(PdfCollection.TILE);
generatedPdf.getCatalog().setCollection(collection);
addAttachment(generatedPdf, filename1, filename1);
addAttachment(generatedPdf, filename2, filename2);
}
}
private static void addAttachment(PdfDocument doc, String attachmentPath, String name) throws IOException {
PdfFileSpec fileSpec = PdfFileSpec.createEmbeddedFileSpec(doc, attachmentPath, name, name, null, null);
doc.addFileAttachment(name, fileSpec);
}
The result in Adobe Acrobat Reader on macOS:

How to change direction of Hebrew letters?

I'm using itext 7.1.8 and I need to save Hebrew text in my document. I found this solution here but it doesn't work for me.
My code looks like the following:
public class RunItextApp {
public static void main(String[] args) throws Exception {
final String filename = "simple.pdf";
final String hebrew = "שדג";
final String text = "\u05E9\u05D3\u05D2";
createSimplePdf(filename, hebrew);
}
private static void createSimplePdf(String filename, String text) throws Exception {
final String path = RunItextApp.class.getResource("/Arial.ttf").getPath();
final PdfFont font = PdfFontFactory.createFont(path, PdfEncodings.IDENTITY_H, true);
Style hebrewStyle = new Style()
.setBaseDirection(BaseDirection.RIGHT_TO_LEFT)
.setTextAlignment(TextAlignment.RIGHT)
.setFontSize(14)
.setFont(font);
final PdfWriter pdfWriter = new PdfWriter(filename);
final PdfDocument pdfDocument = new PdfDocument(pdfWriter);
final Document pdf = new Document(pdfDocument);
pdf.setBaseDirection(BaseDirection.RIGHT_TO_LEFT);
pdf.add(
new Paragraph(text)
.setFontScript(Character.UnicodeScript.HEBREW)
.addStyle(hebrewStyle)
);
pdf.close();
}
}
Why this code doesn't work?
How can I set text direction?
Please take a look at this page https://kb.itextpdf.com/home/it5kb/faq/how-to-set-rtl-direction-for-hebrew-when-converting-html-to-pdf. It is described the same problem you have. I think that the main trick is in text font.

IBM Watson Api Visual Recognition Error

I am trying to implement Watson API for visual recognition. I encounter the following error message:
Here is the code:
public class VisualRecognizer {
private VisualRecognition service;
public VisualRecognizer() {
this.service = new VisualRecognition(VisualRecognition.VERSION_DATE_2016_05_20);
this.service.setApiKey("ourkey");
this.service.setEndPoint("https://gateway-a.watsonplatform.net/visual-recognition/api");
}
public String classifyImage(String filePath) throws FileNotFoundException {
InputStream imageStream = new FileInputStream(filePath);
ClassifyOptions classifyOptions = new ClassifyOptions.Builder()
.imagesFile(imageStream)
.build();
ClassifiedImages result = service.classify(classifyOptions).execute();
Any suggestions on how to solve this would be greatly appreciated.
Please mention a file name.
ClassifyOptions classifyOptions = new ClassifyOptions.Builder()
.imagesFile(imageStream).imagesFilename("xyz.jpg")
.build();
For me it worked

Categories

Resources