def test_explore_pos():
d0, X, y = et.loadboston()
# Adding NA and inf values to dataframe
d0.at[0, 'town'] = None
d0.at[0, 'lon'] = float('inf')
d1 = pd.read_pickle(path + "/data/explore.pkl")
assert et.explore(d0).equals(d1)
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Generated data for pytest
"""
import os.path
path = '/Users/bxp151/ml/000_special_projects/01_exploretransform/exploretransform'
os.chdir(path)
# import inspect, os.path
# filename = inspect.getframeinfo(inspect.currentframe()).filename
# path = os.path.dirname(os.path.abspath(filename))
import exploretransform as et
import plotnine as pn
import pandas as pd
df, X, y = et.loadboston()
# loadboston()
# df.to_pickle(path + "/data/loadboston.pkl", compression = None)
# explore()
# df.at[0, 'town'] = None
# df.at[0,'lon'] = float('inf')
# df = et.explore(df)
# df.to_pickle(path + "/data/explore.pkl", compression = None)
# peek
et.peek(df).to_pickle(path + "/data/peek.pkl", compression=None)
# freq
# f = et.freq(X['rad'])
def test_peek_pos():
d0, X, y = et.loadboston()
d1 = pd.read_pickle(path + "/data/peek.pkl")
assert et.peek(d0).equals(d1)
def test_loadboston_df():
d0, X, y = et.loadboston()
d1 = pd.read_pickle(path + '/data/loadboston.pkl')
assert d0.equals(d1)
def test_CategoricalOtherLevel_pos():
df, d0, y = et.loadboston()
d1 = pd.read_pickle(path + "/data/categoricalotherlevel.pkl")
assert et.CategoricalOtherLevel('town', 0.015).fit_transform(d0).equals(d1)
def test_CorrelationFilter_pos():
df, d0, y = et.loadboston()
d1 = pd.read_pickle(path + "/data/correlationfilter.pkl")
assert et.CorrelationFilter(cut=0.7).fit_transform(d0).equals(d1)
def test_ascores_pos():
df, d0, d1 = et.loadboston()
d0 = d0.select_dtypes('number')
d2 = pd.read_pickle(path + "/data/ascores.pkl")
assert et.ascores(d0, d1).equals(d2)
def test_ColumnSelect_pos():
df, d0, y = et.loadboston()
assert et.ColumnSelect('lon').fit_transform(d0).equals(d0['lon'])
def test_skewstats_pos():
df, d0, y = et.loadboston()
d1 = pd.read_pickle(path + "/data/skewstats.pkl")
assert et.skewstats(d0).equals(d1)
def test_skewstats_nonum():
df, d0, y = et.loadboston()
d0 = d0[['town', 'rad']]
assert et.skewstats(d0) == "Dataframe has no numeric columns"
def test_calcdrop_pos():
df, d0, y = et.loadboston()
d1 = et.corrtable(d0, cut=0.7, full=True)
d2 = et.calcdrop(d1)
d3 = ['nox', 'indus', 'dis', 'crim']
assert set(d2) == set(d3)
def test_corrtable():
df, d0, y = et.loadboston()
d1 = pd.read_pickle(path + "/data/corrtable.pkl")
assert et.corrtable(d0, cut=0.5).equals(d1)
def test_freq():
d0, X, y = et.loadboston()
d1 = pd.read_pickle(path + "/data/freq.pkl")
d2 = et.freq(d0['rad'])
assert d1.equals(d2)