Java Code Examples for org.roaringbitmap.RoaringBitmap#add()

@Setup
public void setup() {
  bitmap1 = new RoaringBitmap();
  bitmap2 = new RoaringBitmap();
  int k = 1 << 16;
  int i = 0;
  for (; i < 10000; ++i) {
    bitmap1.add(i * k);
  }
  for (; i < 10050; ++i) {
    bitmap2.add(i * k);
    bitmap1.add(i * k + 13);
  }
  for (; i < 20000; ++i) {
    bitmap2.add(i * k);
  }
  bitmap1.add(i * k);
}
 

/**
 * @param opt command separated list of Q and C. Q for {@link NumericAttribute}, C for
 *        {@link NominalAttribute}.
 */
@Nonnull
public static RoaringBitmap resolveAttributes(@Nullable final String opt)
        throws UDFArgumentException {
    final RoaringBitmap attr = new RoaringBitmap();
    if (opt == null) {
        return attr;
    }
    final String[] opts = opt.split(",");
    final int size = opts.length;
    for (int i = 0; i < size; i++) {
        final String type = opts[i];
        if ("Q".equals(type)) {
            continue;
        } else if ("C".equals(type)) {
            attr.add(i);
        } else {
            throw new UDFArgumentException("Unsupported attribute type: " + type);
        }
    }
    return attr;
}
 

private void setupEntities(Entities entities) throws Exception {
  if (entities.size() == 0) {
    logger.debug("Skipping initialization of InlinkEntityEntitySimilarity for " + entities.size() + " entities");
    return;
  }

  logger.debug("Initializing InlinkEntityEntitySimilarity for " + entities.size() + " entities");

  entity2vector = new TIntObjectHashMap<>();

  TIntObjectHashMap<int[]> entityInlinks = DataAccess.getInlinkNeighbors(entities);

  for (TIntObjectIterator<int[]> itr = entityInlinks.iterator(); itr.hasNext(); ) {
    itr.advance();
    int entity = itr.key();
    int[] inLinks = itr.value();

    RoaringBitmap bs = new RoaringBitmap();
    for (int l : inLinks) {
      bs.add(l);
    }
    entity2vector.put(entity, bs);
  }

  logger.debug("Done initializing InlinkEntityEntitySimilarity");
}
 

@VisibleForTesting
@Nonnull
public static RoaringBitmap convertAttributeTypes(
        @Nonnull final smile.data.Attribute[] original) {
    final int size = original.length;
    final RoaringBitmap nominalAttrs = new RoaringBitmap();
    for (int i = 0; i < size; i++) {
        smile.data.Attribute o = original[i];
        switch (o.type) {
            case NOMINAL: {
                nominalAttrs.add(i);
                break;
            }
            case NUMERIC: {
                break;
            }
            default:
                throw new UnsupportedOperationException("Unsupported type: " + o.type);
        }
    }
    return nominalAttrs;
}
 

@Setup
public void setup() {
  bitmap1 = new RoaringBitmap();
  bitmap2 = new RoaringBitmap();
  int k = 1 << 16;
  int i = 0;
  for (; i < 10000; ++i) {
    bitmap1.add(i * k);
  }
  for (; i < 10050; ++i) {
    bitmap2.add(i * k);
    bitmap1.add(i * k + 13);
  }
  for (; i < 20000; ++i) {
    bitmap2.add(i * k);
  }
  bitmap1.add(i * k);
}
 

public static void testSuperSparse() {
  System.out.println("Sparse case... universe = [0,"+universe_size+")");
  RoaringBitmap r = new RoaringBitmap();
  int howmany = 100;
  int gap = universe_size / howmany;
  System.out.println("Adding "+howmany+" values separated by gaps of "+gap+ "...");
  System.out.println("As a bitmap it would look like 1000...001000... ");
  for (int i = 1; i < howmany; i++) {
    r.add(i * gap);
  }
  System.out.println("Bits used per value = "+F.format(r.getSizeInBytes()*8.0/howmany));
  r.runOptimize();
  System.out.println("Bits used per value after run optimize = "+F.format(r.getSizeInBytes()*8.0/howmany));
  System.out.println("An uncompressed bitset might use "+F.format(universe_size*1.0/howmany)+" bits per value set");
  System.out.println();

}
 

public static void main(String[] args) {
      RoaringBitmap rr = RoaringBitmap.bitmapOf(1,2,3,1000);
      RoaringBitmap rr2 = new RoaringBitmap();
      rr2.add(4000L,4255L);

      RoaringBitmap rror = RoaringBitmap.or(rr, rr2);// new bitmap
      rr.or(rr2); //in-place computation
      boolean equals = rror.equals(rr);// true
      if(!equals) throw new RuntimeException("bug");
      // number of values stored?
      long cardinality = rr.getLongCardinality();
      System.out.println(cardinality);
      // a "forEach" is faster than this loop, but a loop is possible:
      for(int i : rr) {
        System.out.println(i);
      }
}
 

@Disabled(".add have a different meaning between Roaring64NavigableMap and RoaringBitmap")
@Test
public void testDefaultBehaviorLikeRoaring_MinusOneAsInt() {
  Roaring64NavigableMap longBitmap = newDefaultCtor();
  RoaringBitmap bitmap = new RoaringBitmap();

  longBitmap.addLong(-1);
  bitmap.add(-1);

  // Ok as -1 === -1
  assertEquals(bitmap.select(0), longBitmap.select(0));

  // But RoaringBitmap.select has to be interpreted as an unsigned integer: not as an unsigned
  // long
  assertEquals(Util.toUnsignedLong(bitmap.select(0)), longBitmap.select(0));
}
 

@Test
public void testDefaultBehaviorLikeRoaring() {
  Roaring64NavigableMap longBitmap = newDefaultCtor();
  RoaringBitmap bitmap = new RoaringBitmap();

  longBitmap.addLong(-1);
  bitmap.add(-1);

  longBitmap.addLong(1);
  bitmap.add(1);

  int[] bitmapAsIntArray = bitmap.toArray();

  long[] longBitmapAsArray = longBitmap.toArray();

  // The array seems equivalent, but beware one represents unsigned integers while the others
  // holds unsigned longs: -1 have a different meaning
  assertArrayEquals(bitmapAsIntArray, Ints.toArray(Longs.asList(longBitmapAsArray)));
  assertArrayEquals(Longs.toArray(Ints.asList(bitmapAsIntArray)), longBitmapAsArray);

  long[] bitmapAsLongArray = new long[bitmapAsIntArray.length];
  for (int i = 0; i < bitmapAsIntArray.length; i++) {
    bitmapAsLongArray[i] = Util.toUnsignedLong(bitmapAsIntArray[i]);
  }
}
 

@Test
public void testRoaringBitmap_SelectAboveIntegerMaxValuePlusOne() {
  RoaringBitmap map = new RoaringBitmap();

  long maxForRoaringBitmap = Util.toUnsignedLong(-1) + 1;
  map.add(0L, maxForRoaringBitmap);

  assertEquals(maxForRoaringBitmap, map.getLongCardinality());
  assertEquals(-1, map.select(-1));
}
 

@Test
public void testRoaringBitmap_SelectAboveIntegerMaxValue() {
  RoaringBitmap map = new RoaringBitmap();

  long maxForRoaringBitmap = Util.toUnsignedLong(-1) + 1;
  map.add(0L, maxForRoaringBitmap);

  assertEquals(maxForRoaringBitmap, map.getLongCardinality());
  assertEquals(-1, map.select(-1));
}
 

public static void testAlternating() {
  System.out.println("Alternating case... universe = [0,"+universe_size+")");
  RoaringBitmap r = new RoaringBitmap();
  for (int i = 1; i < universe_size; i++) {
    if(i%2 == 0)
      r.add(i);
  }
  System.out.println("Adding all even values in the universe");
  System.out.println("As a bitmap it would look like 01010101... ");
  System.out.println("Bits used per value = "+F.format(r.getSizeInBytes()*8.0/r.getCardinality()));
  r.runOptimize();
  System.out.println("Bits used per value after run optimize = "+F.format(r.getSizeInBytes()*8.0/r.getCardinality()));
  System.out.println("An uncompressed bitset might use "+F.format(universe_size*1.0/r.getCardinality())+" bits per value set");
  System.out.println();
}
 

@Setup
public void setup() {
  bitmap1 = new RoaringBitmap();
  bitmap2 = new RoaringBitmap();
  int k = 1 << 16;
  for (int i = 0; i < 10000; ++i) {
    bitmap1.add(i * k);
    bitmap2.add(i * k);
  }
}
 

@Setup
public void setup() {
  bitmap1 = new RoaringBitmap();
  bitmap2 = new RoaringBitmap();
  int k = 1 << 16;
  for (int i = 0; i < 10000; ++i) {
    bitmap1.add(i * k);
    bitmap2.add(i * k);
  }
}
 

@Override
public void insertToBand(int band, int[] hashes, int id) {
    String hexkey = toHex(band, hashes);
    RoaringBitmap idsBitmap = maps.get(hexkey);
    if (idsBitmap == null) {
        idsBitmap = RoaringBitmap.bitmapOf(id);
        maps.put(hexkey, idsBitmap);
    } else {
        idsBitmap.add(id);
    }
}
 

private RoaringBitmap toBitmap(Selection otherSelection) {
  if (otherSelection instanceof BitmapBackedSelection) {
    return ((BitmapBackedSelection) otherSelection).bitmap.clone();
  }
  RoaringBitmap bits = new RoaringBitmap();
  for (int i : otherSelection) {
    bits.add(i);
  }
  return bits;
}
 

private RoaringBitmap toBitmap(Selection otherSelection) {
  if (otherSelection instanceof BitmapBackedSelection) {
    return ((BitmapBackedSelection) otherSelection).bitmap.clone();
  }
  RoaringBitmap bits = new RoaringBitmap();
  for (int i : otherSelection) {
    bits.add(i);
  }
  return bits;
}
 

@Setup
public void setup() {
  bitmap1 = new RoaringBitmap();
  bitmap2 = new RoaringBitmap();
  int k = 1 << 16;
  for (int i = 0; i < 10000; ++i) {
    bitmap1.add(2 * i * k);
    bitmap2.add(2 * i * k + 1);
  }
}
 

@Setup
public void setup() {
  bitmap1 = new RoaringBitmap();
  bitmap2 = new RoaringBitmap();
  int k = 1 << 16;
  for (int i = 0; i < 10000; ++i) {
    bitmap1.add(2 * i * k);
    bitmap2.add(2 * i * k + 1);
  }
}
 

@Test
public void testTitanicPruning() throws IOException, ParseException {
    String datasetUrl =
            "https://gist.githubusercontent.com/myui/7cd82c443db84ba7e7add1523d0247a9/raw/f2d3e3051b0292577e8c01a1759edabaa95c5781/titanic_train.tsv";

    URL url = new URL(datasetUrl);
    InputStream is = new BufferedInputStream(url.openStream());

    DelimitedTextParser parser = new DelimitedTextParser();
    parser.setColumnNames(true);
    parser.setDelimiter(",");
    parser.setResponseIndex(new NominalAttribute("survived"), 0);

    AttributeDataset train = parser.parse("titanic train", is);
    double[][] x_ = train.toArray(new double[train.size()][]);
    int[] y = train.toArray(new int[train.size()]);

    // pclass, name, sex, age, sibsp, parch, ticket, fare, cabin, embarked
    // C,C,C,Q,Q,Q,C,Q,C,C
    RoaringBitmap nominalAttrs = new RoaringBitmap();
    nominalAttrs.add(0);
    nominalAttrs.add(1);
    nominalAttrs.add(2);
    nominalAttrs.add(6);
    nominalAttrs.add(8);
    nominalAttrs.add(9);

    int columns = x_[0].length;
    Matrix x = new RowMajorDenseMatrix2d(x_, columns);
    int numVars = (int) Math.ceil(Math.sqrt(columns));
    int maxDepth = Integer.MAX_VALUE;
    int maxLeafs = Integer.MAX_VALUE;
    int minSplits = 2;
    int minLeafSize = 1;
    int[] samples = null;
    PRNG rand = RandomNumberGeneratorFactory.createPRNG(43L);

    final String[] featureNames = new String[] {"pclass", "name", "sex", "age", "sibsp",
            "parch", "ticket", "fare", "cabin", "embarked"};
    final String[] classNames = new String[] {"yes", "no"};
    DecisionTree tree = new DecisionTree(nominalAttrs, x, y, numVars, maxDepth, maxLeafs,
        minSplits, minLeafSize, samples, SplitRule.GINI, rand) {
        @Override
        public String toString() {
            return predictJsCodegen(featureNames, classNames);
        }
    };
    tree.toString();
}