def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
rowSum = 0
for j in range(colCount):
if BASE == 2:
# 50/50
# r = h2o_util.choice_with_probability([(0, .5), (1, .5)])
# 98/2
r = h2o_util.choice_with_probability([(0, .98), (1, .2)])
else:
raise Exception("Unsupported BASE: " + BASE)
rowSum += r
rowData.append(r)
responseVar = rowSum % BASE
# make r a many-digit real, so gzip compresses even more better!
rowData.append('%#034.32e' % responseVar)
rowDataCsv = ",".join(map(str, rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
rowSum = 0
for j in range(colCount):
if BASE == 2:
# 50/50
# r = h2o_util.choice_with_probability([(0, .5), (1, .5)])
# 98/2
r = h2o_util.choice_with_probability([(0, 0.98), (1, 0.2)])
else:
raise Exception("Unsupported BASE: " + BASE)
rowSum += r
rowData.append(r)
responseVar = rowSum % BASE
# make r a many-digit real, so gzip compresses even more better!
rowData.append("%#034.32e" % responseVar)
rowDataCsv = ",".join(map(str, rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generatators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
for j in range(colCount):
r = h2o_util.choice_with_probability([(1.1, .05), (0.1, .95)])
rowData.append(r)
rowDataCsv = ",".join(map(str, rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generatators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
for j in range(colCount):
r = h2o_util.choice_with_probability([(1.1, .05), (0.1, .95)])
rowData.append(r)
rowDataCsv = ",".join(map(str,rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generatators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
for j in range(colCount):
# r = h2o_util.choice_with_probability([(1.1, .02), (0.1, .98)])
r = h2o_util.choice_with_probability([(1, ONE_RATE), (0, 1 - ONE_RATE)])
# make r a many-digit real, so gzip compresses even more better!
# rowData.append('%#034.32e' % r)
rowData.append('%.1f' % r)
rowDataCsv = ",".join(map(str,rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generatators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
for j in range(colCount):
# r = h2o_util.choice_with_probability([(1.1, .02), (0.1, .98)])
r = h2o_util.choice_with_probability([(1, ONE_RATE), (0, 1 - ONE_RATE)])
# make r a many-digit real, so gzip compresses even more better!
# rowData.append('%#034.32e' % r)
rowData.append('%.1f' % r)
rowDataCsv = ",".join(map(str,rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
rowSum = 0
for j in range(colCount):
if BASE==2:
# we're just doing 50/50 for now, unlike the print says above
r = h2o_util.choice_with_probability([(0, .5), (1, .5)])
else:
raise Exception("Unsupported BASE: " + BASE)
rowSum += r
rowData.append(r)
responseVar = rowSum % BASE
rowData.append(responseVar)
rowDataCsv = ",".join(map(str,rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def write_syn_dataset(csvPathname, rowCount, colCount, SEED):
# 8 random generators, 1 per column
r1 = random.Random(SEED)
dsf = open(csvPathname, "w+")
for i in range(rowCount):
rowData = []
rowSum = 0
for j in range(colCount):
if BASE == 2:
# we're just doing 50/50 for now, unlike the print says above
r = h2o_util.choice_with_probability([(0, .5), (1, .5)])
else:
raise Exception("Unsupported BASE: " + BASE)
rowSum += r
rowData.append(r)
responseVar = rowSum % BASE
rowData.append(responseVar)
rowDataCsv = ",".join(map(str, rowData))
dsf.write(rowDataCsv + "\n")
dsf.close()
def good_choices(n):
ch = h2o_util.choice_with_probability([(ch1,0.1), (ch2,0.1), (ch3,0.1), (ch4,0.1), (ch5,0.6)])
return ch(n)
