def test_inverse_transform(self):
from lale.lib.sklearn import OneHotEncoder, OrdinalEncoder
fproc_ohe = OneHotEncoder(handle_unknown="ignore")
# test_init_fit_transform
trained_ohe = fproc_ohe.fit(self.X_train, self.y_train)
transformed_X = trained_ohe.transform(self.X_test)
orig_X_ohe = trained_ohe._impl._wrapped_model.inverse_transform(transformed_X)
fproc_oe = OrdinalEncoder(handle_unknown="ignore")
# test_init_fit_transform
trained_oe = fproc_oe.fit(self.X_train, self.y_train)
transformed_X = trained_oe.transform(self.X_test)
orig_X_oe = trained_oe._impl.inverse_transform(transformed_X)
self.assertEqual(orig_X_ohe.all(), orig_X_oe.all())
def test_bool_label(self):
import pandas as pd
data_records = [
{'IS_TENT':False, 'GENDER':'M', 'AGE':20, 'MARITAL_STATUS':'Single', 'PROFESSION':'Sales'},
{'IS_TENT':False, 'GENDER':'M', 'AGE':20, 'MARITAL_STATUS':'Single', 'PROFESSION':'Sales'},
{'IS_TENT':False, 'GENDER':'F', 'AGE':37, 'MARITAL_STATUS':'Single', 'PROFESSION':'Other'},
{'IS_TENT':False, 'GENDER':'M', 'AGE':42, 'MARITAL_STATUS':'Married', 'PROFESSION':'Other'},
{'IS_TENT':True, 'GENDER':'F', 'AGE':24, 'MARITAL_STATUS':'Married', 'PROFESSION':'Retail'},
{'IS_TENT':False, 'GENDER':'F', 'AGE':24, 'MARITAL_STATUS':'Married', 'PROFESSION':'Retail'},
{'IS_TENT':False, 'GENDER':'M', 'AGE':29, 'MARITAL_STATUS':'Single', 'PROFESSION':'Retail'},
{'IS_TENT':False, 'GENDER':'M', 'AGE':29, 'MARITAL_STATUS':'Single', 'PROFESSION':'Retail'},
{'IS_TENT':True, 'GENDER':'M', 'AGE':43, 'MARITAL_STATUS':'Married', 'PROFESSION':'Trades'},
{'IS_TENT':False, 'GENDER':'M', 'AGE':43, 'MARITAL_STATUS':'Married', 'PROFESSION':'Trades'}]
df = pd.DataFrame.from_records(data_records)
X = df.drop(['IS_TENT'], axis=1).values
y = df['IS_TENT'].values
from lale.lib.sklearn import OneHotEncoder as Enc
from lale.lib.sklearn import GradientBoostingClassifier as Clf
trainable = Enc() >> Clf()
trained = trainable.fit(X, y)
def test_bool_label(self):
import pandas as pd
data_records = [
{
"IS_TENT": False,
"GENDER": "M",
"AGE": 20,
"MARITAL_STATUS": "Single",
"PROFESSION": "Sales",
},
{
"IS_TENT": False,
"GENDER": "M",
"AGE": 20,
"MARITAL_STATUS": "Single",
"PROFESSION": "Sales",
},
{
"IS_TENT": False,
"GENDER": "F",
"AGE": 37,
"MARITAL_STATUS": "Single",
"PROFESSION": "Other",
},
{
"IS_TENT": False,
"GENDER": "M",
"AGE": 42,
"MARITAL_STATUS": "Married",
"PROFESSION": "Other",
},
{
"IS_TENT": True,
"GENDER": "F",
"AGE": 24,
"MARITAL_STATUS": "Married",
"PROFESSION": "Retail",
},
{
"IS_TENT": False,
"GENDER": "F",
"AGE": 24,
"MARITAL_STATUS": "Married",
"PROFESSION": "Retail",
},
{
"IS_TENT": False,
"GENDER": "M",
"AGE": 29,
"MARITAL_STATUS": "Single",
"PROFESSION": "Retail",
},
{
"IS_TENT": False,
"GENDER": "M",
"AGE": 29,
"MARITAL_STATUS": "Single",
"PROFESSION": "Retail",
},
{
"IS_TENT": True,
"GENDER": "M",
"AGE": 43,
"MARITAL_STATUS": "Married",
"PROFESSION": "Trades",
},
{
"IS_TENT": False,
"GENDER": "M",
"AGE": 43,
"MARITAL_STATUS": "Married",
"PROFESSION": "Trades",
},
]
df = pd.DataFrame.from_records(data_records)
X = df.drop(["IS_TENT"], axis=1).values
y = df["IS_TENT"].values
from lale.lib.sklearn import GradientBoostingClassifier as Clf
from lale.lib.sklearn import OneHotEncoder as Enc
trainable = Enc() >> Clf()
_ = trainable.fit(X, y)