mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-28 18:42:26 +00:00
81 lines
2.7 KiB
Python
81 lines
2.7 KiB
Python
#!/usr/bin/env python3
|
|
import os
|
|
import sys
|
|
|
|
import numpy as np
|
|
import pandas as pd
|
|
from scipy import stats
|
|
|
|
CURDIR = os.path.dirname(os.path.realpath(__file__))
|
|
sys.path.insert(0, os.path.join(CURDIR, "helpers"))
|
|
|
|
from pure_http_client import ClickHouseClient
|
|
|
|
|
|
def test_and_check(name, a, b, t_stat, p_value):
|
|
client = ClickHouseClient()
|
|
client.query("DROP TABLE IF EXISTS mann_whitney;")
|
|
client.query(
|
|
"CREATE TABLE mann_whitney (left Float64, right UInt8) ENGINE = Memory;"
|
|
)
|
|
client.query(
|
|
"INSERT INTO mann_whitney VALUES {};".format(
|
|
", ".join(["({},{})".format(i, 0) for i in a])
|
|
)
|
|
)
|
|
|
|
client.query(
|
|
"INSERT INTO mann_whitney VALUES {};".format(
|
|
", ".join(["({},{})".format(i, 1) for i in b])
|
|
)
|
|
)
|
|
|
|
real = client.query_return_df(
|
|
"SELECT roundBankers({}(left, right).1, 16) as t_stat, ".format(name)
|
|
+ "roundBankers({}(left, right).2, 16) as p_value ".format(name)
|
|
+ "FROM mann_whitney FORMAT TabSeparatedWithNames;"
|
|
)
|
|
real_t_stat = real["t_stat"][0]
|
|
real_p_value = real["p_value"][0]
|
|
assert abs(
|
|
real_t_stat - np.float64(t_stat) < 1e-2
|
|
), "clickhouse_t_stat {}, scipy_t_stat {}".format(real_t_stat, t_stat)
|
|
assert (
|
|
abs(real_p_value - np.float64(p_value)) < 1e-2
|
|
), "clickhouse_p_value {}, scipy_p_value {}".format(real_p_value, p_value)
|
|
client.query("DROP TABLE IF EXISTS mann_whitney;")
|
|
|
|
|
|
def test_mann_whitney():
|
|
rvs1 = np.round(stats.norm.rvs(loc=1, scale=5, size=500), 5)
|
|
rvs2 = np.round(stats.expon.rvs(scale=0.2, size=500), 5)
|
|
s, p = stats.mannwhitneyu(rvs1, rvs2, alternative="two-sided")
|
|
test_and_check("mannWhitneyUTest", rvs1, rvs2, s, p)
|
|
test_and_check("mannWhitneyUTest('two-sided')", rvs1, rvs2, s, p)
|
|
|
|
equal = np.round(stats.cauchy.rvs(scale=5, size=500), 5)
|
|
s, p = stats.mannwhitneyu(equal, equal, alternative="two-sided")
|
|
test_and_check("mannWhitneyUTest('two-sided')", equal, equal, s, p)
|
|
|
|
s, p = stats.mannwhitneyu(equal, equal, alternative="less", use_continuity=False)
|
|
test_and_check("mannWhitneyUTest('less', 0)", equal, equal, s, p)
|
|
|
|
rvs1 = np.round(stats.cauchy.rvs(scale=10, size=65536), 5)
|
|
rvs2 = np.round(stats.norm.rvs(loc=0, scale=10, size=65536), 5)
|
|
s, p = stats.mannwhitneyu(rvs1, rvs2, alternative="greater")
|
|
test_and_check("mannWhitneyUTest('greater')", rvs1, rvs2, s, p)
|
|
|
|
|
|
def test_mann_whitney_skew():
|
|
rvs1 = [1]
|
|
rvs2 = [0, 2, 4]
|
|
s, p = stats.mannwhitneyu(rvs1, rvs2, alternative="two-sided")
|
|
test_and_check("mannWhitneyUTest", rvs1, rvs2, s, p)
|
|
test_and_check("mannWhitneyUTest('two-sided')", rvs1, rvs2, s, p)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
test_mann_whitney()
|
|
test_mann_whitney_skew()
|
|
print("Ok.")
|