Examples with Val water.rapids.Val used on opensource projects

Example 6 with Val

use of water.rapids.Val in project h2o-3 by h2oai.

the class AstRBind method apply.

@Override
public ValFrame apply(Env env, Env.StackHelp stk, AstRoot[] asts) {
    // Execute all args.  Find a canonical frame; all Frames must look like this one.
    // Each argument turns into either a Frame (whose rows are entirely
    // inlined) or a scalar (which is replicated across as a single row).
    // Canonical Frame; all frames have the same column count, types and names
    Frame fr = null;
    // Total chunks
    int nchks = 0;
    // Computed AstRoot results
    Val[] vals = new Val[asts.length];
    for (int i = 1; i < asts.length; i++) {
        vals[i] = stk.track(asts[i].exec(env));
        if (vals[i].isFrame()) {
            fr = vals[i].getFrame();
            // Total chunks
            nchks += fr.anyVec().nChunks();
        } else
            // One chunk per scalar
            nchks++;
    }
    // No Frame, just a pile-o-scalars?
    // The zero-length vec for the zero-frame frame
    Vec zz = null;
    if (fr == null) {
        // Zero-length, 1-column, default name
        fr = new Frame(new String[] { Frame.defaultColName(0) }, new Vec[] { zz = Vec.makeZero(0) });
        if (asts.length == 1)
            return new ValFrame(fr);
    }
    // Verify all Frames are the same columns, names, and types.  Domains can vary, and will be the union
    // Input frame
    final Frame[] frs = new Frame[asts.length];
    // Column types
    final byte[] types = fr.types();
    // Compute a new layout!
    final long[] espc = new long[nchks + 1];
    int coffset = 0;
    Frame[] tmp_frs = new Frame[asts.length];
    for (int i = 1; i < asts.length; i++) {
        // Save values computed for pass 2
        Val val = vals[i];
        Frame fr0 = val.isFrame() ? val.getFrame() : // Scalar: auto-expand into a 1-row frame
        (tmp_frs[i] = new Frame(fr._names, Vec.makeCons(val.getNum(), 1L, fr.numCols())));
        // Check that all frames are compatible
        if (fr.numCols() != fr0.numCols())
            throw new IllegalArgumentException("rbind frames must have all the same columns, found " + fr.numCols() + " and " + fr0.numCols() + " columns.");
        if (!Arrays.deepEquals(fr._names, fr0._names))
            throw new IllegalArgumentException("rbind frames must have all the same column names, found " + Arrays.toString(fr._names) + " and " + Arrays.toString(fr0._names));
        if (!Arrays.equals(types, fr0.types()))
            throw new IllegalArgumentException("rbind frames must have all the same column types, found " + Arrays.toString(types) + " and " + Arrays.toString(fr0.types()));
        // Save frame
        frs[i] = fr0;
        // Roll up the ESPC row counts
        long roffset = espc[coffset];
        long[] espc2 = fr0.anyVec().espc();
        for (// Roll up the row counts
        int j = 1; // Roll up the row counts
        j < espc2.length; // Roll up the row counts
        j++) espc[coffset + j] = (roffset + espc2[j]);
        // Chunk offset
        coffset += espc2.length - 1;
    }
    if (zz != null)
        zz.remove();
    // build up the new domains for each vec
    HashMap<String, Integer>[] dmap = new HashMap[types.length];
    String[][] domains = new String[types.length][];
    int[][][] cmaps = new int[types.length][][];
    for (int k = 0; k < types.length; ++k) {
        dmap[k] = new HashMap<>();
        int c = 0;
        byte t = types[k];
        if (t == Vec.T_CAT) {
            int[][] maps = new int[frs.length][];
            for (int i = 1; i < frs.length; i++) {
                maps[i] = new int[frs[i].vec(k).domain().length];
                for (int j = 0; j < maps[i].length; j++) {
                    String s = frs[i].vec(k).domain()[j];
                    if (!dmap[k].containsKey(s))
                        dmap[k].put(s, maps[i][j] = c++);
                    else
                        maps[i][j] = dmap[k].get(s);
                }
            }
            cmaps[k] = maps;
        } else {
            cmaps[k] = new int[frs.length][];
        }
        domains[k] = c == 0 ? null : new String[c];
        for (Map.Entry<String, Integer> e : dmap[k].entrySet()) domains[k][e.getValue()] = e.getKey();
    }
    // Now make Keys for the new Vecs
    Key<Vec>[] keys = fr.anyVec().group().addVecs(fr.numCols());
    Vec[] vecs = new Vec[fr.numCols()];
    int rowLayout = Vec.ESPC.rowLayout(keys[0], espc);
    for (int i = 0; i < vecs.length; i++) vecs[i] = new Vec(keys[i], rowLayout, domains[i], types[i]);
    // Do the row-binds column-by-column.
    // Switch to F/J thread for continuations
    AstRBind.ParallelRbinds t;
    H2O.submitTask(t = new AstRBind.ParallelRbinds(frs, espc, vecs, cmaps)).join();
    for (Frame tfr : tmp_frs) if (tfr != null)
        tfr.delete();
    return new ValFrame(new Frame(fr.names(), t._vecs));
}

Example 7 with Val

use of water.rapids.Val in project h2o-3 by h2oai.

the class StratifiedSplitTest method testStratifiedSampling.

@Test
public void testStratifiedSampling() {
    f = ArrayUtils.frame("response", vec(ari(1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1)));
    fanimal = ArrayUtils.frame("response", vec(ar("dog", "cat"), ari(1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1)));
    f = new Frame(f);
    fanimal = new Frame(fanimal);
    f._key = Key.make();
    fanimal._key = Key.make();
    DKV.put(f);
    DKV.put(fanimal);
    //
    Val res1 = Rapids.exec("(h2o.random_stratified_split (cols_py " + f._key + " 0) 0.3333333 123)");
    fr1 = res1.getFrame();
    // minority class should be in the test split
    Assert.assertEquals(fr1.vec(0).at8(0), 1);
    // minority class should be in the train split
    Assert.assertEquals(fr1.vec(0).at8(11), 0);
    // minority class should be in the train split
    Assert.assertEquals(fr1.vec(0).mean(), 1.0 / 3.0, 1e-5);
    //test categorical
    //
    Val res2 = Rapids.exec("(h2o.random_stratified_split (cols_py " + fanimal._key + " 0) 0.3333333 123)");
    fr2 = res2.getFrame();
    // minority class should be in the test split
    Assert.assertEquals(fr2.vec(0).at8(0), 1);
    // minority class should be in the test split
    Assert.assertEquals(fr2.vec(0).at8(11), 0);
    // minority class should be in the test split
    Assert.assertEquals(fr2.vec(0).mean(), 1.0 / 3.0, 1e-5);
}

Example 8 with Val

use of water.rapids.Val in project h2o-3 by h2oai.

the class AstRectangleConditionalAssignTest method testConditionalAssignCategorical.

@Test
public void testConditionalAssignCategorical() {
    Frame fr = makeTestFrame();
    Vec expected = cvec(new String[] { "a", "b" }, "b", "b", "b", "b", "b");
    try {
        Val val = Rapids.exec("(tmp= py_1 (:= data \"b\" 4 (== (cols_py data 4) \"a\")))");
        if (val instanceof ValFrame) {
            Frame fr2 = val.getFrame();
            assertCatVecEquals(expected, fr2.vec(4));
            fr2.remove();
        }
    } finally {
        fr.remove();
        expected.remove();
    }
}

Example 9 with Val

use of water.rapids.Val in project h2o-3 by h2oai.

the class AstRectangleConditionalAssignTest method testConditionalAssignString.

@Test
public void testConditionalAssignString() {
    Frame fr = makeTestFrame();
    Vec expected = svec("row1", "tst", "row3", "tst", "row5");
    try {
        Val val = Rapids.exec("(tmp= py_1 (:= data \"tst\" 3 (== (cols_py data 4) \"a\")))");
        if (val instanceof ValFrame) {
            Frame fr2 = val.getFrame();
            assertStringVecEquals(expected, fr2.vec(3));
            fr2.remove();
        }
    } finally {
        fr.remove();
        expected.remove();
    }
}

Example 10 with Val

use of water.rapids.Val in project h2o-3 by h2oai.

the class AstRectangleFrameSliceAssignTest method testAssignFrameSlice.

@Test
public void testAssignFrameSlice() throws Exception {
    final Frame data = parse_test_file(Key.make("data"), "smalldata/airlines/allyears2k_headers.zip");
    Frame output = null;
    try {
        String rapids = "(tmp= tst (:= data (rows (cols data [8.0, 11.0] ) [10000.0:" + _nRows + ".0] ) [8.0, 11.0] [0.0:" + _nRows + ".0] ) )";
        Val val = Rapids.exec(rapids);
        if (val instanceof ValFrame) {
            output = val.getFrame();
            // categorical column
            String[] expectedCats = catVec2array(data.vec(8));
            System.arraycopy(expectedCats, 10000, expectedCats, 0, _nRows);
            String[] actualCats = catVec2array(output.vec(8));
            assertArrayEquals(expectedCats, actualCats);
            // numerical column
            double[] expected = vec2array(data.vec(11));
            System.arraycopy(expected, 10000, expected, 0, _nRows);
            double[] actual = vec2array(output.vec(11));
            assertArrayEquals(expected, actual, 0.0001d);
        }
    } finally {
        data.delete();
        if (output != null) {
            output.delete();
        }
    }
}