def transform(self, df, **transform_params):
"""
Returns a copy of ``df`` holding the columns defined in ``cols``
"""
# TODO add a test that self.cols is a subset of the columns if df
pdu._is_cols_subset_of_df_cols(self.cols, df)
return df[self.cols].copy()
def fit(self, df, y=None, **fit_params):
"""
Doesn't do anything.
Provided for the sake of consistency with scikit-learn.
"""
pdu._is_cols_subset_of_df_cols(self.cols, df)
return self
def transform(self, df, **transform_params):
"""
Label encoding ``cols`` of ``df`` using the fitting
Parameters
----------
df : DataFrame
DataFrame to be preprocessed
"""
if not self._is_fitted:
raise NotFittedError("Fitting was not preformed")
pdu._is_cols_subset_of_df_cols(self.cols, df)
df = df.copy()
label_enc_dict = {}
for col in self.cols:
label_enc_dict[col] = self.les[col].transform(df[col])
labelenc_cols = pd.DataFrame(
label_enc_dict,
# The index of the resulting DataFrame should be assigned and
# equal to the one of the original DataFrame. Otherwise, upon
# concatenation NaNs will be introduced.
index=df.index)
for col in self.cols:
df[col] = labelenc_cols[col]
return df
def transform(self, df, **transform_params):
"""
Scaling ``cols`` of ``df`` using the fitting
Parameters
----------
df : DataFrame
DataFrame to be preprocessed
"""
if not self._is_fitted:
raise NotFittedError("Fitting was not preformed")
pdu._is_cols_subset_of_df_cols(self.cols, df)
df = df.copy()
standartize_cols = pd.DataFrame(
# StandardScaler returns a NumPy.array, and thus indexing
# breaks. Explicitly fixed next.
self.standard_scaler.transform(df[self.cols]),
columns=self.cols,
# The index of the resulting DataFrame should be assigned and
# equal to the one of the original DataFrame. Otherwise, upon
# concatenation NaNs will be introduced.
index=df.index)
df = df.drop(self.cols, axis=1)
df = pd.concat([df, standartize_cols], axis=1)
return df
def transform(self, df, **transform_params):
"""
Returns a copy of ``df`` with a new column (named ``feat_name``) and
holding the day of the week as derived from ``df.col``.
"""
df = df.copy()
if isinstance(df, pd.DataFrame):
pdu._is_cols_subset_of_df_cols([self.col], df)
df[self.feat_name] = df[self.col].apply(
lambda x: x.dayofweek).astype('category')
else:
raise ValueError("Non supported input")
return df
def transform(self, df, **transform_params):
"""
Compute the ratio between ``df[col]`` and the mean/median which
was fitted.
"""
check_is_fitted(self, 'const_')
df = df.copy()
if isinstance(df, pd.DataFrame):
pdu._is_cols_subset_of_df_cols([self.col], df)
df[self.feat_name] = df[self.col].div(self.const_)
else:
raise ValueError("Non supported input")
return df
def fit(self, df, y=None, **fit_params):
"""
Fitting the preprocessing
Parameters
----------
df : DataFrame
Data to use for fitting.
In many cases, should be ``X_train``.
"""
pdu._is_cols_subset_of_df_cols(self.cols, df)
self.standard_scaler.fit(df[self.cols])
self._is_fitted = True
return self
def fit(self, df, y=None, **fit_params):
"""
Fits the instance to the mean/median of ``df[col]``
"""
if not isinstance(df, pd.DataFrame):
raise ValueError("Non supported input")
pdu._is_cols_subset_of_df_cols([self.col], df)
if self.func == 'mean':
self.const_ = df[self.col].mean()
elif self.func == 'median':
self.const_ = df[self.col].median()
else:
raise ValueError(
"Unsupported function ({}). Can be either mean or "
"median".format(self.func))
return self
def fit(self, df, y=None, **fit_params):
"""
Fitting the instance on ``df``
Parameters
----------
df : DataFrame
The base DataFrame from which column names are learned.
These names will be used when transforming data
"""
pdu._is_cols_subset_of_df_cols(self.cols, df)
self.ohe.fit(df[self.cols])
self.ohe_cols_names_ = []
for col in self.cols:
for i in range(self.n_values):
self.ohe_cols_names_.append(col + '_' + str(i))
return self