// covtype.altered response column has this distribution:
// -1 20510
// 1 211840
// 2 283301
// 3 35754
// 4 2747
// 6 17367
// 10000 9493
@Test
public void testSplitCats() {
Frame cov = parse_test_file(Key.make("cov"), "smalldata/covtype/covtype.altered.gz");
System.out.println(cov.toString(0, 10));
Val v_ddply = Exec.exec("(ddply cov [54] nrow)");
System.out.println(v_ddply.toString());
((ValFrame) v_ddply)._fr.delete();
Val v_groupby = Exec.exec("(GB cov [54] nrow 54 \"all\")");
System.out.println(v_groupby.toString());
((ValFrame) v_groupby)._fr.delete();
cov.delete();
}
// ==========================================================================
/*@Test*/ public void testBasicCRUD() {
// Parse a file with many broken enum/string columns
Key k = Key.make("zip.hex");
try {
Frame fr = TestUtil.parseFrame(k, "smalldata/zip_code/zip_code_database.csv.gz");
System.out.println(fr);
StringBuilder sb = new StringBuilder();
String[] fs = fr.toStringHdr(sb);
int lim = Math.min(40, (int) fr.numRows());
for (int i = 0; i < lim; i++) fr.toString(sb, fs, i);
System.out.println(sb.toString());
} finally {
UKV.remove(k);
}
}
@Test
public void testGroupbyTableSpeed() {
Frame ids = parse_test_file(Key.make("cov"), "smalldata/junit/id_cols.csv");
ids.replace(0, ids.anyVec().toCategoricalVec()).remove();
System.out.println(ids.toString(0, 10));
long start = System.currentTimeMillis();
Val v_gb = Exec.exec("(GB cov [0] nrow 0 \"all\")");
System.out.println("GB Time= " + (System.currentTimeMillis() - start) + "msec");
System.out.println(v_gb.toString());
((ValFrame) v_gb)._fr.delete();
long start2 = System.currentTimeMillis();
Val v_tb = Exec.exec("(table cov FALSE)");
System.out.println("Table Time= " + (System.currentTimeMillis() - start2) + "msec");
System.out.println(v_tb.toString());
((ValFrame) v_tb)._fr.delete();
ids.delete();
}
@Test
public void testCovtype() {
Frame frame = parse_test_file("smalldata/covtype/covtype.20k.data");
AggregatorModel.AggregatorParameters parms = new AggregatorModel.AggregatorParameters();
parms._train = frame._key;
parms._radius_scale = 5.0;
long start = System.currentTimeMillis();
AggregatorModel agg = new Aggregator(parms).trainModel().get(); // 0.179
System.out.println(
"AggregatorModel finished in: "
+ (System.currentTimeMillis() - start) / 1000.
+ " seconds");
agg.checkConsistency();
frame.delete();
Frame output = agg._output._output_frame.get();
Log.info("Exemplars: " + output.toString());
output.remove();
Log.info("Number of exemplars: " + agg._exemplars.length);
// Assert.assertTrue(agg._exemplars.length==615);
agg.remove();
}
@Override
public String toString() {
return _fr.toString();
}
public String toStringAll() {
StringBuilder sb = new StringBuilder();
String[] fs = toStringHdr(sb);
for (int i = 0; i < numRows(); i++) toString(sb, fs, i);
return sb.toString();
}