def label_encode_engineer(self):
# must be called AFTER sg_ordinals
lce = LabelCountEncoder()
self.label_df = self.ord_df.copy()
for c in self.train().columns:
if self.label_df[c].dtype == 'object':
lce = LabelCountEncoder()
self.label_df[c] = lce.fit_transform(self.label_df[c])
def label_encode_engineer(self):
lce = LabelCountEncoder()
for c in self.all.columns:
if self.all[c].dtype == 'object':
lce = LabelCountEncoder()
self.all[c] = lce.fit_transform(self.all[c])
dfS=df[['SalePrice','Heating','HeatingQC', 'CentralAir', 'Electrical', '1stFlrSF', \
'2ndFlrSF', \
'LowQualFinSF', 'GrLivArea', 'BsmtFullBath', 'BsmtHalfBath', 'FullBath', \
'HalfBath', 'BedroomAbvGr', 'KitchenAbvGr', 'KitchenQual', \
'TotRmsAbvGrd', 'Functional', 'Fireplaces', 'FireplaceQu']]
dfS[['GrLivArea', '1stFlrSF']].corr()
dfS.GrLivArea.value_counts()
dfS[['GrLivArea', '2ndFlrSF']].corr()
dfS.BsmtHalfBath.value_counts()
# Wonder if the Bsmt bathrooms parameters covary with BsmtCond or BsmtFinType1
# %% Heating Variables
from LabelClass import LabelCountEncoder
lce = LabelCountEncoder()
dfS['Heating'] = lce.fit_transform(dfS['Heating'])
dfS[['HeatingQC', 'Heating']].corr()
df.HeatingQC.value_counts()
df.Heating.value_counts()
# %%
df.loc[df.Functional == 'Maj2']
dfS.loc[dfS.Functional == 3]
dfS['Kitchen*Quality'] = dfS.KitchenAbvGr * dfS.KitchenQual
functional_dic = {
'Typ': 8,
'Min1': 7,