def test_output():
# check if the outputs are correct.
#test value
assert train_encode1.feature_cat_chr.iloc[
0] == 0.43, 'The encoded value for training dataset is wrong'
assert np.isclose(
test_encode2['feature_cat_chr'].iloc[0],
0.5) == True, 'The encoded value for unseen test dataset is wrong'
#check shape
assert train_encode1.shape == train1.shape, "The shape of training dataset is wrong"
assert test_encode1.shape == test1.shape, "The shape of testing datset is wrong"
#check when X_test is none
assert len(
target_encoder.target_encoder(
X_train=train1,
y=target_cha,
cat_columns=['feature_cat_chr', 'feature_cat_num'],
objective='binary')) == 1
def check_exception():
# check if the function handles invalid inputs.
# check input of objective
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=train1,
y=train1.target_bin,
cat_columns=['feature_cat_chr', 'feature_cat_num'],
X_test=test1,
prior=0.5,
objective='something')
# check if cat_columns is a list
with pytest.raises(Exception):
target_encoder.target_encoder(X_train=train1,
y=train1.target_bin,
cat_columns="not list")
# check if prior is a numeric value
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=train1,
y=train1.target_bin,
cat_columns=['feature_cat_chr', 'feature_cat_num'],
prior='string')
# check if y is a pandas series
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=train1,
y=[1, 2],
cat_columns=['feature_cat_chr', 'feature_cat_num'])
# check if length y equals to length X_train
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=train1,
y=pd.Series([1, 2]),
cat_columns=['feature_cat_chr', 'feature_cat_num'])
# check if X_train is pandas dataframe
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=[1, 2],
y=train1.target_bin,
cat_columns=['feature_cat_chr', 'feature_cat_num'])
# check if X_train contains cat_columns
with pytest.raises(Exception):
target_encoder.target_encoder(X_train=train1,
y=train1.target_bin,
cat_columns=['something'])
# check if target variable is numeric for regression objective
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=train1,
y=target_cha,
cat_columns=['feature_cat_chr', 'feature_cat_num'])
# check if target is binary
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=train1,
y=train1.target_cont,
cat_columns=['feature_cat_chr', 'feature_cat_num'],
objective='binary')
# check if X_test is pandas dataframe
with pytest.raises(Exception):
target_encoder.target_encoder(
X_train=train1,
y=train1.target_cont,
cat_columns=['feature_cat_chr', 'feature_cat_num'],
X_test=[1, 2])
# check if X_test contains cat_columns
with pytest.raises(Exception):
target_encoder.target_encoder(X_train=train1,
y=train1.target_cont,
cat_columns=['something'],
X_test=test1)
from encoderpy import target_encoder
import pandas as pd
import pytest
data = pd.read_csv("data/testing_data.csv")
train1 = data.query("train_test_1 == 'train'")
test1 = data.query("train_test_1 == 'test'")
train2 = data.query("train_test_3 == 'train'")
test2 = data.query("train_test_3 == 'test'")
train_encode1, test_encode1 = target_encoder.target_encoder(
X_train=train1,
y=train1.target_bin,
cat_columns=['feature_cat_chr', 'feature_cat_num'],
X_test=test1,
prior=0.5,
objective='binary')
train_encode2, test_encode2 = target_encoder.target_encoder(
X_train=train2,
y=train2.target_bin,
cat_columns=['feature_cat_chr', 'feature_cat_num'],
X_test=test2,
prior=0.5,
objective='binary')
target_cha = train1.target_bin.replace({
train1.target_bin.unique()[0]: "a",
train1.target_bin.unique()[1]: "b"