def categoricals(self,
model_name='onehot_model.pkl',
cols=None,
owr=False,
model_bin=None):
"""Onehot encoder on categoricals."""
self.log('Apply onehot encoder on categorical')
model_path = os.path.join(self.model_path, model_name)
if cols is None:
cols = self.data.cat_cols
if ((not os.path.isfile(model_path)) or owr) and (model_bin is None):
self.log('\nTrain model\n')
model_bin = OneHotEncoder(
cols=cols,
use_cat_names=True,
handle_unknown='error',
drop_invariant=False,
impute_missing=False)
model_bin.fit(self.data._X)
self.data._X = model_bin.transform(self.data._X)
setattr(model_bin, 'data_schema', self.data._X.columns.values)
# Save model
if self.auto_save:
joblib.dump(model_bin, model_path)
elif os.path.isfile(model_path):
# File exists/prediction:
model_bin = joblib.load(model_path)
self.data._X = model_bin.transform(self.data._X)
self.data.check_schema(model_bin, '_X')
else:
# Prediction in pipeline
self.data._X = model_bin.transform(self.data._X)
self.data.check_schema(model_bin, '_X')
return model_bin
def _encode_categories(self):
"""
This private method stands for encoding categorical variables. Label encoding used for ordinal categories and
one-hot encoding used for nominal categories.
"""
logging.info(f'#{self._index()} - Encoding categorical columns...')
# get column names for categorical and numerical features
categorical_vars = self.X.select_dtypes(include='object').columns
numerical_vars = self.X.columns.difference(categorical_vars)
ordinal = pd.Index([
'ExterQual', 'ExterCond', 'BsmtQual', 'BsmtCond', 'HeatingQC',
'KitchenQual', 'FireplaceQu', 'GarageQual', 'GarageCond', 'PoolQC'
])
nominal = categorical_vars.difference(ordinal)
standard_mapping = {
'NA': 0,
'Po': 1,
'Fa': 2,
'TA': 3,
'Gd': 4,
'Ex': 5
}
mapping_for_ordinals = [{
'col': column,
'mapping': standard_mapping
} for column in ordinal]
x_num = self.X[numerical_vars]
x_test_num = self.X_test[numerical_vars]
# one hot encode categorical columns
one_hot_encoder = OneHotEncoder(use_cat_names=True)
label_encoder = OrdinalEncoder(drop_invariant=True,
mapping=mapping_for_ordinals,
handle_unknown='error')
x_cat_nom = one_hot_encoder.fit_transform(self.X[nominal])
x_cat_ord = label_encoder.fit_transform(self.X[ordinal])
x_test_cat_nom = one_hot_encoder.transform(self.X_test[nominal])
x_test_cat_ord = label_encoder.transform(self.X_test[ordinal])
self.X = x_num.join(x_cat_ord).join(x_cat_nom)
self.X_test = x_test_num.join(x_test_cat_ord).join(x_test_cat_nom)
logging.info(f'#{self._step_index} - DONE!')
def fit_onehot(input_df: pd.DataFrame, cols: List[str], na_value: Any = None):
"""
Creates the One-hot encoder by fitting it through the given DataFrame
NaN values and Special value specified under `na_value` in the DataFrame will be encoded as unseen value.
Args:
input_df: DataFrame used to fit the encoder
cols: List of categorical columns to be encoded
na_value: Default null value for DataFrame
Returns:
result_df: encoded input_df DataFrame
model : encoder model to be passed to `transform_onehot` method
"""
df = input_df.copy()
if na_value is not None:
for col in cols:
df[col] = df[col].replace({na_value: np.nan})
drop_cols = ["{}_nan".format(col) for col in cols]
encoder = OneHotEncoder(cols=cols, use_cat_names=True)
encoder = encoder.fit(df)
result_df = encoder.transform(df)
for drop_col in drop_cols:
if drop_col in result_df.columns:
result_df = result_df.drop(columns=[drop_col])
model = {
"encoder": encoder,
"cols": cols,
"na_value": na_value,
"drop_cols": drop_cols,
}
return result_df, model
def encode_low_cardinality_categorical_df(dataframe, fit=False):
"""
Encode low cardinality categorical features using OneHot Encoding and dropping invariant features
---
Arguments
dataframe: pd.DataFrame
Dataframe with pre-processed data (i.e. renamed features), low card. categorical features only
fit: boolean
Indicates if we should train or load an encoder
Returns
dataframe: pd.DataFrame
Dataframe with encoded data
"""
# Train or load an encoder
if fit:
encoder = OneHotEncoder(cols=dataframe.columns.values, drop_invariant=True)
encoder.fit(dataframe)
pickle_obj(encoder, 'low_card_categorical_encoder')
else:
encoder = unpickle_obj('low_card_categorical_encoder')
# transform data
return encoder.transform(dataframe)
'PassengerId', 'Survived']] ########################################################################### # Split data into train and test # ########################################################################### trainData = fullData.loc[fullData.DataPartition == 'train'] testData = fullData.loc[fullData.DataPartition == 'test'] ########################################################################### # One hot encode # ########################################################################### # https://github.com/scikit-learn-contrib/categorical-encoding # http://contrib.scikit-learn.org/categorical-encoding/onehot.html from category_encoders import OneHotEncoder categories = list(set(trainData.select_dtypes(['category']).columns)) target = trainData.Survived enc = OneHotEncoder(cols=categories,return_df = 1, handle_unknown = 'ignore').fit(trainData, target) trainData = enc.transform(trainData) testData = enc.transform(testData) ########################################################################### # Drop multi collinear levels and no longer required # ########################################################################### dropColumns = ['DataPartition'] trainData = trainData.drop(columns=dropColumns) testData = testData.drop(columns=dropColumns) testData = testData.drop(columns='Survived') ########################################################################### # Start h2o cloud # ########################################################################### import h2o h2o.init() h2o.remove_all # clean slate, in case cluster was already running # upload data to h2o cloud
from category_encoders import OneHotEncoder encoder = OneHotEncoder(cols=['color'], use_cat_names=True) train = encoder.fit_transform(train) test = encoder.fit_transform(test) train.head() from sklearn.preprocessing import LabelEncoder encoder = LabelEncoder() encoder.fit(train['type']) print(encoder.classes_) train['type_no'] = encoder.transform(train['type']) train.head() sns.heatmap(train.corr(), xticklabels=list(train), yticklabels=list(train)) target = train['type_no'] # for visualizations target_string = train['type'] # for final predictions del train['type'] del train['type_no'] target.head() from sklearn.model_selection import train_test_split train_data, test_data, train_target, test_target = train_test_split(train, target, test_size=0.2, random_state=42) from mlxtend.plotting import plot_decision_regions import matplotlib.gridspec as gridspec import itertools
class RFEncoder(BaseEstimator, TransformerMixin):
def __init__(self,
cols=None,
handle_missing='value',
handle_unknown='value',
use_cat_names=False,
return_df=True,
max_subsets=None,
max_depth=3,
n_estimators=100,
min_count=1,
n_jobs=1):
self.cols = cols
self.handle_missing = handle_missing
self.handle_unknown = handle_unknown
self.use_cat_names = use_cat_names
self.return_df = return_df
self.max_subsets = max_subsets
self.max_depth = max_depth
self.n_estimators = n_estimators
self.n_jobs = n_jobs
self.min_count = min_count
def fit(self, X, y=None):
self._dim = X.shape[1]
if self.cols is None:
self.cols = get_obj_cols(X)
self.dummy_encoder = OneHotEncoder(cols=self.cols,
handle_unknown='value',
handle_missing='value')
self.dummy_encoder = self.dummy_encoder.fit(X)
self.mapping = self.generate_mapping(X, y)
X_temp = self.transform(X, override_return_df=True)
self.feature_names = list(X_temp.columns)
return self
def generate_mapping(self, X, y):
X = self.dummy_encoder.transform(X.copy(deep=True))
y = y.copy(deep=True)
mapping = []
for switch in self.dummy_encoder.mapping:
col = switch.get('col')
values = switch.get('mapping').copy(deep=True)
if isinstance(self.max_depth, int):
max_depth = self.max_depth
elif isinstance(self.max_depth, float):
max_depth = round(self.max_depth * values.shape[1])
else:
max_depth = min(self.max_depth[1],
round(self.max_depth[0] * values.shape[1]))
if max_depth == 0:
continue
forest = RandomForestClassifier(
max_depth=max_depth,
n_estimators=self.n_estimators,
n_jobs=self.n_jobs,
)
forest.fit(X[values.columns], y)
subsets = self.get_subsets(forest.decision_path(values))
subset_df = pd.DataFrame(data=subsets,
index=values.index,
columns=[
'{col}_subset_{i}'.format(col=col,
i=i)
for i in range(subsets.shape[1])
])
base_df = values.join(subset_df)
mapping.append({'col': col, 'mapping': base_df})
return mapping
def get_subsets(self, decision_path):
subset_sizes = np.asarray(decision_path[0].sum(axis=0))[0]
subsets = decision_path[0][:, subset_sizes != 1].toarray()
subsets, count = np.unique(subsets, return_counts=True, axis=1)
subsets = subsets[:, count >= self.min_count]
count = count[count >= self.min_count]
subsets = subsets[:, np.argsort(-count)]
subset_sizes = subsets.sum(axis=0)
subsets = subsets[:, np.argsort(subset_sizes)]
if self.max_subsets is not None:
subsets = subsets[:, :self.max_subsets]
return subsets
def transform(self, X, override_return_df=False):
if self.handle_missing == 'error':
if X[self.cols].isnull().any().any():
raise ValueError('Columns to be encoded can not contain null')
if self._dim is None:
raise ValueError(
'Must train encoder before it can be used to transform data.')
if X.shape[1] != self._dim:
raise ValueError('Unexpected input dimension %d, expected %d' % (
X.shape[1],
self._dim,
))
if not list(self.cols):
return X if self.return_df else X.values
X = self.dummy_encoder.ordinal_encoder.transform(X)
if self.handle_unknown == 'error':
if X[self.cols].isin([-1]).any().any():
raise ValueError(
'Columns to be encoded can not contain new values')
X = self.get_dummies(X)
if self.return_df or override_return_df:
return X
else:
return X.values
def get_dummies(self, X_in):
X = X_in.copy(deep=True)
cols = X.columns.values.tolist()
for switch in self.mapping:
col = switch.get('col')
mod = switch.get('mapping')
base_df = mod.reindex(X[col])
base_df = base_df.set_index(X.index)
X = pd.concat([base_df, X], axis=1)
old_column_index = cols.index(col)
cols[old_column_index:old_column_index + 1] = mod.columns
X = X.reindex(columns=cols)
return X
def get_feature_names(self):
if not isinstance(self.feature_names, list):
raise ValueError(
'Must transform data first. Affected feature names are not known before.'
)
else:
return self.feature_names
def _encode_categories(self):
"""
This private method stands for encoding categorical variables. Label encoding used for ordinal categories and
one-hot encoding used for nominal categories.
"""
logging.info(f'#{self._index()} - Encoding categorical columns...')
def encode(data):
# encode Sex column
data['Sex'] = data['Sex'] == 'male'
# encode Name column
name_cols = data['Name'].apply(lambda x: pd.Series(
[str(x).split(",")[0],
str(x).split(", ")[1].split(".")[0]],
index=['Family name', 'Title']))
data = data.join(name_cols)
# identify Titles with same meaning
data['Title'].replace({
'Mlle': 'Miss',
'Ms': 'Miss',
'Mme': 'Mrs'
},
inplace=True)
# group rare Titles
title_names = (data['Title'].value_counts() < 10)
data['Title'] = data['Title'].apply(lambda x: 'Misc'
if title_names.loc[x] else x)
# create Family size and Alone column from SibSp, Parch cols
data['Family size'] = data['SibSp'] + data['Parch'] + 1
data['Alone'] = data['Family size'] == 1
# make 5 equal size groups from Fares
data['Fare'] = pd.qcut(data['Fare'], 5, labels=False)
# make 5 groups from Ages
data['Age'] = pd.cut(data['Age'], 5, labels=False)
# rename columns and delete unnecessary features
data = data.rename(columns={
'Sex': 'Male',
'Fare': 'FareBins',
'Age': 'AgeBins'
})
data.drop(['Name', 'SibSp', 'Parch'], axis=1, inplace=True)
return data
self.X = encode(self.X)
self.X_test = encode(self.X_test)
for col in self.X.columns:
if self.X[col].dtype != 'float64':
table = self.X.join(self.y)[[col, 'Survived'
]].groupby(col,
as_index=False).mean()
table['Survived'] = (table['Survived'] * 100).map(
'{:.2f} %'.format)
logging.info(
f'Survival ratio by: {col}\n{table}\n{"-" * 10}\n')
one_hot_encoder = OneHotEncoder(use_cat_names=True)
one_hot_columns = one_hot_encoder.fit_transform(
self.X[['Title', 'Embarked']])
one_hot_columns_test = one_hot_encoder.transform(
self.X_test[['Title', 'Embarked']])
self.X = self.X.join(one_hot_columns)
self.X_test = self.X_test.join(one_hot_columns_test)
self.X.drop(['Family name', 'Title', 'Embarked'], axis=1, inplace=True)
self.X_test.drop(['Family name', 'Title', 'Embarked'],
axis=1,
inplace=True)
logging.info(f'#{self._step_index} - DONE!')
# StandardScaler - pre-processor to put numerical column in the same scale
scaler = StandardScaler().fit(x_train)
scaler
values_scale = scaler.transform(x_train)
values_scale[:10]
x_train = scaler.transform(x_train)
# generate the model - could be any model
# instance of the classifier decision tree and train the model
clf_tree = tree.DecisionTreeClassifier()
clf_tree = clf_tree.fit(x_train, y_train)
# Apply object ohe and pre-processor on data for test
x_test = ohe.transform(x_test)
scaler_test = StandardScaler().fit(x_test)
x_test = scaler_test.transform(x_test)
x_test[:10]
# predict
clf_tree.predict(x_test)
# Validate the model
acuracy = clf_tree.score(x_test, y_test)
acuracy
# Pipeline
# will create a kind of alias for each method
pip_1 = Pipeline([('ohe', OneHotEncoder()), ('scaler', StandardScaler()),
X.head() # In[249]: one_hot= OneHotEncoder(cols=["user_name","country","hint_variety"],use_cat_names=True) # OneHotEncoder from category_encoders package one_hot.fit(X) # In[250]: X=one_hot.transform(X) # In[253]: X.columns # #### Word2vec implementation - # In[259]: wine2vec.build_vocab(review_desc.values)
df[['host_is_superhost', 'bathrooms_text', 'has_availability', 'instant_bookable']].astype(float)
df.head
pip install category_encoders
# Instantiate transformer - one hot encoder
from category_encoders import OneHotEncoder
transformer = OneHotEncoder(use_cat_names=True)
# Transform to fit training data
transformer.fit(df)
# Transform our training data
df = transformer.transform(df)
X = df.drop('price', axis=1)
y= df['price']
X = X.astype(float)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
from keras.layers import BatchNormalization, Dropout
import keras
model = Sequential([
# The Input Layer :
Dense(1024, input_dim = X_train.shape[1]),
# Define col names for the parameters of the network
pred_vars = ['MONTH', 'ORIGIN', 'DEST', 'DISTANCE']
target_var = 'PASSENGERS'
keep = pred_vars
keep.append(target_var)
# Subset only what's needed
data = data[keep]
# Encode the source and target nodes using a catagory encoder
from category_encoders import OneHotEncoder
ce = OneHotEncoder()
ce.fit(data)
# transform the encoded data
data_encoded = ce.transform(data)
labels = data[target_var]
data_encoded.drop(target_var, 1, inplace=True)
# split out a final eval set
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(data_encoded,
labels,
random_state=0,
test_size=.25)
# convert to xgb data format
import xgboost as xgb
dtrain = xgb.DMatrix(X_train, label=y_train)
dtest = xgb.DMatrix(X_test, label=y_test)
]]
###########################################################################
# Split data into train and test #
###########################################################################
trainData = fullData.loc[fullData.DataPartition == 'train']
testData = fullData.loc[fullData.DataPartition == 'test']
###########################################################################
# One hot encode #
###########################################################################
# https://github.com/scikit-learn-contrib/categorical-encoding
# http://contrib.scikit-learn.org/categorical-encoding/onehot.html
categories = list(set(trainData.select_dtypes(['category']).columns))
target = trainData.Survived
enc = OneHotEncoder(cols=categories, return_df=1,
handle_unknown='ignore').fit(trainData, target)
trainData = enc.transform(trainData)
testData = enc.transform(testData)
###########################################################################
# Drop multi collinear levels and no longer required #
###########################################################################
dropColumns = ['DataPartition']
trainData = trainData.drop(columns=dropColumns)
testData = testData.drop(columns=dropColumns)
testData = testData.drop(columns='Survived')
###########################################################################
# Start h2o cloud #
###########################################################################
h2o.init()
h2o.remove_all # clean slate, in case cluster was already running
# upload data to h2o cloud
train = h2o.H2OFrame(trainData)
return score
######### Creating objects for 2 classification models.
logit = LogisticRegression(random_state=SEED)
rf = RandomForestClassifier(random_state=SEED)
###################################################################################################
######### Apply One Hot Encoding
from category_encoders import OneHotEncoder
onehot_enc = OneHotEncoder(cols=X_Columns)
onehot_enc.fit(X_train, y_train)
print('Original number of features: \n', X_train.shape[1], "\n")
data_ohe_train = onehot_enc.fit_transform(X_train)
data_ohe_test = onehot_enc.transform(X_test)
print('Features after OHE: \n', data_ohe_train.shape[1])
######### Logistic Regression
onehot_logit_score = get_score(logit, data_ohe_train, y_train, data_ohe_test,
y_test)
print('Logistic Regression score with One hot encoding:', onehot_logit_score)
######### Random Forest
onehot_rf_score = get_score(rf, data_ohe_train, y_train, data_ohe_test, y_test)
print('Random Forest score with One hot encoding:', onehot_logit_score)
###################################################################################################
######### Apply Hashing Encoding
from category_encoders import HashingEncoder
hashing_enc = HashingEncoder(n_components=10000, cols=X_Columns)
class Encoder():
encode_methods = {
'OrdinalEncoder': OrdinalEncoder,
'OneHotEncoder': OneHotEncoder,
'CountEncoder': CountEncoder,
'TargetEncoder': TargetEncoder,
}
# spark_encode_methods = {
# 'mean_encoder':,
# 'target_encoder':,
# 'label_encoder':,
# 'onehot_encoder'
# }
# target_encoder，mean_encoder在编码时，不能够把训练集和验证机concat在一起进行编码
# label_encoder,onehot_encoder可以
def __init__(self,
sparksess=None,
logdir='/encoder',
handle_unknown='-99999',
save_encoder=False):
self.spark = sparksess
self.logdir = logdir
self.save_encoder
self.ordinal_encoder_features = []
self.onehot_encoder_features = []
self.count_encoder_features = []
self.target_encoder_features = []
self.ordinal_encoder = OrdinalEncoder(
cols=self.ordinal_encoder_features,
return_df=True,
handle_unknown=handle_unknown)
self.onehot_encoder = OneHotEncoder(cols=self.onehot_encoder_features,
return_df=True,
handle_unknown=handle_unknown)
self.count_encoder = CountEncoder(cols=self.count_encoder_features,
return_df=True,
handle_unknown=handle_unknown)
self.target_encoder = TargetEncoder(cols=self.target_encoder_features,
return_df=True,
handle_unknown=handle_unknown)
def fit(self,
x_train,
x_val=None,
y_train=None,
y_val=None,
method_mapper=None):
"""
Parameters
----------
x_train: pd.DataFrame
x_val: pd.DataFrame
y_train: pd.DataFrame
y_val: pd.DataFrame
method_mapper: dict
a mapping of feature to EncodeMethod
example mapping:
{
'feature1': OrdinalEncoder,
'feature2': OneHotEncoder,
'feature3': CountEncoder,
'feature4': TargetEncoder,
}
"""
for feat in method_mapper:
if method_mapper[feat] == 'OrdinalEncoder':
self.ordinal_encoder_features.append(feat)
elif method_mapper[feat] == 'OneHotEncoder':
self.onehot_encoder_features.append(feat)
elif method_mapper[feat] == 'CountEncoder':
self.count_encoder_features.append(feat)
elif method_mapper[feat] == 'TargetEncoder':
self.target_encoder_features.append(feat)
else:
raise ValueError(
'编码方式只支持[OrdinalEncoder, OneHotEncoder, CountEncoder, TargetEncoder], 接收到%s'
% feat)
if self.spark is None:
if len(self.ordinal_encoder_features) != 0 or len(
self.onehot_encoder_features) != 0:
x_whole = x_train.append(x_val)
y_whole = None
if not y_train is None and not y_val is None:
y_whole = y_train.append(y_val)
x_whole = self.ordinal_encoder.fit_transform(x_whole, y_whole)
x_whole = self.onehot_encoder.fit_transform(x_whole, y_whole)
x_train = x_whole[:len(x_train)]
x_val = x_whole[len(x_train):]
x_train = self.count_encoder.fit_transform(x_train, y_train)
x_val = self.count_encoder.transform(x_val, y_val)
x_train = self.target_encoder.fit_transform(x_train, y_train)
x_val = self.target_encoder.transform(x_val, y_val)
if self.save_encoder:
self.save_encoder()
return x_train, y_train, x_val, y_val
def transform(self, x, y=None):
x = self.ordinal_encoder.transform(x, y)
x = self.onehot_encoder.transform(x, y)
x = self.count_encoder.transform(x, y)
x = self.target_encoder.transform(x, y)
return x, y
def fit_transform(self,
x_train,
x_val=None,
y_train=None,
y_val=None,
method_mapper=None):
"""
Parameters
----------
x_train: pd.DataFrame
x_val: pd.DataFrame
y_train: pd.DataFrame
y_val: pd.DataFrame
method_mapper: dict
a mapping of feature to EncodeMethod
example mapping:
{
'feature1': OrdinalEncoder,
'feature2': OneHotEncoder,
'feature3': CountEncoder,
'feature4': TargetEncoder,
}
"""
self.fit(x_train, x_val, y_train, y_val, method_mapper)
x_train, y_train = self.transform(x_train, y_train)
if x_val is not None:
x_val, y_val = self.transform(x_val, y_val)
return x_train, y_train, x_val, y_val
def save_encoder(self):
now = time.strftime("%Y-%m-%d-%H_%M_%S", time.localtime(time.time()))
os.makedirs(os.path.join(self.logdir, now))
with open(os.path.join(self.logdir, now, 'OrdinalEncoder.pkl'),
'wb') as f:
pickle.dump(self.ordinal_encoder, f)
with open(os.path.join(self.logdir, now, 'OneHotEncoder.pkl'),
'wb') as f:
pickle.dump(self.onehot_encoder, f)
with open(os.path.join(self.logdir, now, 'CountEncoder.pkl'),
'wb') as f:
pickle.dump(self.count_encoder, f)
with open(os.path.join(self.logdir, now, 'TargetEncoder.pkl'),
'wb') as f:
pickle.dump(self.target_encoder, f)
with open(
os.path.join(self.logdir, now, 'OrdinalEncoderFeatures.json'),
'w') as f:
json.dump(self.ordinal_encoder_features, f)
with open(os.path.join(self.logdir, now, 'OneHotEncoderFeatures.json'),
'w') as f:
json.dump(self.onehot_encoder_features, f)
with open(os.path.join(self.logdir, now, 'CountEncoderFeatures.json'),
'w') as f:
json.dump(self.count_encoder_features, f)
with open(os.path.join(self.logdir, now, 'TargetEncoderFeatures.json'),
'w') as f:
json.dump(self.target_encoder_features, f)
def load_encoder(self, logdir=None):
with open(os.path.join(self.logdir, 'OrdinalEncoder.pkl'), 'wb') as f:
pickle.dump(self.ordinal_encoder, f)
with open(os.path.join(self.logdir, 'OneHotEncoder.pkl'), 'wb') as f:
pickle.dump(self.onehot_encoder, f)
with open(os.path.join(self.logdir, 'CountEncoder.pkl'), 'wb') as f:
pickle.dump(self.count_encoder, f)
with open(os.path.join(self.logdir, 'TargetEncoder.pkl'), 'wb') as f:
pickle.dump(self.target_encoder, f)
with open(os.path.join(self.logdir, 'OrdinalEncoderFeatures.json'),
'w') as f:
json.dump(self.ordinal_encoder_features, f)
with open(os.path.join(self.logdir, 'OneHotEncoderFeatures.json'),
'w') as f:
json.dump(self.onehot_encoder_features, f)
with open(os.path.join(self.logdir, 'CountEncoderFeatures.json'),
'w') as f:
json.dump(self.count_encoder_features, f)
with open(os.path.join(self.logdir, 'TargetEncoderFeatures.json'),
'w') as f:
json.dump(self.target_encoder_features, f)