Python extract_cabin_letter Examples

Example #1

File: score.py Project: quezadasubiabre/DeploymentMachineLeatning

def predict(data):

    # extract first letter from cabin

    pf.extract_cabin_letter(data, 'cabin')

    # impute NA categorical
    for var in ['age', 'fare']:
        pf.add_missing_indicator(data, var)

    # impute NA numerical

    for var in config.CATEGORICAL_VARS:
        pf.impute_na(data, var)

    # Group rare labels
    for var in config.CATEGORICAL_VARS:
        pf.remove_rare_labels(data, var, config.FREQUENT_LABELS)

    # encode variables
    data = pf.encode_categorical(df, config.CATEGORICAL_VARS)

    # scale variables

    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions, _ = pf.predict(data, config.OUTPUT_MODEL_PATH)

    return predictions

Example #2

File: score.py Project: annplaube/mldeploy

def predict(data):

    # extract first letter from cabin
    pf.extract_cabin_letter(data, 'cabin')

    # impute NA numerical
    for var in config.NUMERICAL_TO_IMPUTE:
        pf.add_missing_indicator(data, var)
        pf.impute_na(data, var, config.IMPUTATION_DICT[var])

    # impute NA categorical
    for var in config.CATEGORICAL_VARS:
        pf.impute_na(data, var)

    # Group rare labels
    for var, labels in config.FREQUENT_LABELS.items():
        pf.remove_rare_labels(data, var, labels)

    # encode variables
    for var in config.CATEGORICAL_VARS:
        data = pf.encode_categorical(data, var)

    # check all dummies were added
    pf.check_dummy_variables(data, config.DUMMY_VARIABLES)

    data = data[config.ALL_VARS]

    # scale variables
    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)

    return predictions

Example #3

File: score.py Project: wonglimey/machine-learning-pipeline

def predict(data):

    # extract first letter from cabin
    data[config.EXTRACT_VARIABLE] = pf.extract_cabin_letter(
        data, config.EXTRACT_VARIABLE)

    # impute NA categorical
    for var in config.CATEGORICAL_TO_ENCODE:
        data[var] = pf.impute_na(data, var, replacement='Missing')

    # impute NA numerical
    for var in config.NUMERICAL_TO_IMPUTE:
        if (var == 'age'):
            data[var] = pf.add_missing_indicator(data, var, config.AGE_MEDIAN)
        else:
            data[var] = pf.add_missing_indicator(data, var, config.FARE_MEDIAN)

    # Group rare labels
    for var in config.CATEGORICAL_TO_ENCODE:
        data[var] = pf.remove_rare_labels(data, var, config.RARE_VALUE)

    # encode variables
    for var in config.CATEGORICAL_TO_ENCODE:
        data = pf.encode_categorical(data, var)

    # check all dummies were added
    pf.check_dummy_variables(data, config.DUMMY_VARIABLE)

    # scale variables
    data = pf.scale_features(data[config.FEATURES], config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)

    return predictions

Example #4

File: score.py Project: overdrivek/MLDeployment

def predict(data):
    
    # extract first letter from cabin
    X_test = pf.extract_cabin_letter(data, config.IMPUTATION_DICT['cabin_variable'])

    # impute NA categorical
    X_test = pf.add_missing_indicator(X_test, config.CATEGORICAL_VARS)
    
    
    # impute NA numerical
    for var in config.NUMERICAL_TO_IMPUTE:
        X_test = pf.impute_na(X_test,var,replace_by=config.IMPUTATION_DICT[var], add_na_columns=True)

    
    # Group rare labels
    X_test = pf.remove_rare_labels(X_test, config.FREQUENT_LABELS)
    
    # encode variables
    for var in config.CATEGORICAL_VARS:
        X_test = pf.encode_categorical(X_test, var)
    X_test.drop(labels=config.CATEGORICAL_VARS, axis=1, inplace=True)
        
    # check all dummies were added
    X_test = pf.check_dummy_variables(X_test, config.DUMMY_VARIABLES)

    
    # scale variables
    X_test = pf.scale_features(X_test, config.OUTPUT_SCALER_PATH)
    
    # make predictions
    predictions = pf.predict(X_test,config.OUTPUT_MODEL_PATH)

    
    return predictions

Example #5

File: score.py Project: Vodolazskyi/deployment-ml-models-course

def predict(data):

    # extract first letter from cabin
    data["cabin"] = pf.extract_cabin_letter(data, "cabin")

    # impute NA categorical
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.impute_na(data, var, replacement="Missing")

    # impute NA numerical
    for var in config.NUMERICAL_TO_IMPUTE:
        median_val = data[var].median()
        data[var] = pf.impute_na(data, var, replacement=median_val)

    # Group rare labels
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.remove_rare_labels(data, var, config.FREQUENT_LABELS[var])

    # encode variables
    for var in config.CATEGORICAL_VARS:
        data = pf.encode_categorical(data, var)

    # check all dummies were added
    data = pf.check_dummy_variables(data, config.DUMMY_VARIABLES)

    # scale variables
    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)
    return predictions

Example #6

def predict(data):

    # extract first letter from cabin
    data['cabin'] = pf.extract_cabin_letter(data, 'cabin')

    # impute NA categorical
    data[config.CATEGORICAL_VARS] = pf.impute_na(data[config.CATEGORICAL_VARS],
                                                 'Missing')

    # impute NA numerical
    data[config.NUMERICAL_TO_IMPUTE] = pf.impute_na(
        data[config.NUMERICAL_TO_IMPUTE], 'Numerical')

    # Group rare labels
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.remove_rare_labels(data, var,
                                          config.FREQUENT_LABELS[var])

    # encode variables
    data = pf.encode_categorical(data, config.CATEGORICAL_VARS)
    print(data.shape)

    # check all dummies were added
    data = pf.check_dummy_variables(data, config.DUMMY_VARIABLES)
    print(data.shape)

    # scale variables
    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)

    return predictions

Example #7

File: score.py Project: rmania/ml_deployment

def predict(data):
    
    # extract first letter from cabin
    data['cabin'] = pf.extract_cabin_letter(data, var='cabin')

    # impute NA categorical
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.impute_na(data, var, replacement='Missing')
    
    
    # impute NA numerical
    for var in config.NUMERICAL_TO_IMPUTE:
        data[var] = pf.add_missing_indicator(data, var)
    
    
    # Group rare labels
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.remove_rare_labels(data, var, frequent_labels = config.FREQUENT_LABELS[var])
    
    # encode variables
    data = pf.encode_categorical(data, var=config.CATEGORICAL_VARS)
        
        
    # check all dummies were added
    data = pf.check_dummy_variables(data, dummy_list = config.DUMMY_VARIABLES)
    
    # scale variables
    data = pf.scale_features(data[config.FEATURES],
                             config.OUTPUT_SCALER_PATH)
    
    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)

    
    return predictions

Example #8

File: score.py Project: nimecan/Deployment-of-ML-Models

def predict(data):

    # extract first letter from cabin
    data["cabin"] = pf.extract_cabin_letter(data, "cabin")

    # impute NA categorical
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.impute_na(data, var, replacement='Missing')

    # impute NA numerical
    for var in config.NUMERICAL_TO_IMPUTE:
        data[var + "_na"] = pf.add_missing_indicator(data, var)
        median_train_var = config.IMPUTATION_DICT[var]
        data[var] = pf.impute_na(data, var, replacement=median_train_var)

    # Group rare labels
    for var in config.CATEGORICAL_VARS:
        freq_labels = config.FREQUENT_LABELS[var]
        # Remove rare labels from both train and test set
        data[var] = pf.remove_rare_labels(data, var, freq_labels)

    # encode variables
    for var in config.CATEGORICAL_VARS:
        data = pf.encode_categorical(data, var)

    # check all dummies were added
    data = pf.check_dummy_variables(data, config.DUMMY_VARIABLES)

    # scale variables
    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)

    return predictions

Example #9

def predict(data):
    
    # extract first letter from cabin
    data = pf.extract_cabin_letter(data, 'cabin')


    # impute NA categorical
    for var in config.CATEGORICAL_VARS:
        data = pf.impute_na(data, var, config.IMPUTATION_DICT)    
    
    # impute NA numerical
    for var in ['age', 'fare']:
        data = pf.impute_na(data, var, config.IMPUTATION_DICT)
    
    # add indicator variables
    for var in ['age', 'fare']:
        data = pf.add_missing_indicator(data, var)
    
    
    # Group rare labels
    for var in config.CATEGORICAL_VARS:
        data = pf.remove_rare_labels(data, config.FREQUENT_LABELS, var)
    
    # encode variables
    for var in config.CATEGORICAL_VARS:
        data = pf.encode_categorical(data, var)
        
        
    # check all dummies were added
    data = pf.check_dummy_variables(data, config.DUMMY_VARIABLES)

    
    # scale variables
    data = pf.scale_features(data, config.ORDERED_COLUMNS, config.OUTPUT_SCALER_PATH)

    
    # make predictions
    predictions = pf.predict(data, config.ORDERED_COLUMNS, config.OUTPUT_MODEL_PATH)

    
    return predictions

Example #10

def predict(data):

    data = pf.load_data(config.PATH_TO_DATASET)

    X_train, X_test, y_train, y_test = pf.divide_train_test(data, config.TARGET)
    data = X_test.copy()

        # impute categorical variables
    data = pf.add_missing_indicator(data, config.CATEGORICAL_VARS)

    # extract first letter from cabin
    data = pf.extract_cabin_letter(data, 'cabin')

    # impute NA categorical
    data = pf.impute_na(data, config.CATEGORICAL_VARS)

    # impute NA numerical
    data = pf.add_missing_indicator(data, config.NUMERICAL_TO_IMPUTE)
    data = pf.impute_num(data, config.NUMERICAL_TO_IMPUTE)

    # Group rare labels
    data = pf.remove_rare_labels(data, config.CATEGORICAL_VARS)

    # encode variables
    data, data_features = pf.encode_categorical(data, config.CATEGORICAL_VARS)

    print(data.head(1))
    
    # check all dummies were added
    data = pf.check_dummy_variables(data, config.DUMMY_VARIABLES)
    
    # scale variables
    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    class_, pred = pf.predict(data, config.OUTPUT_MODEL_PATH)

    
    return class_

Example #11

def predict(data):

    # extract first letter from cabin
    data['cabin'] = pf.extract_cabin_letter(data, 'cabin')

    # impute NA categorical
    for el in config.CATEGORICAL_VARS:
        data[el] = pf.impute_na(data, el, replacement='Missing')

    # impute NA numerical
    for el in config.NUMERICAL_TO_IMPUTE:

        # add missing indicator first
        data[el + '_NA'] = pf.add_missing_indicator(data, el)

        # impute NA
        data[el] = pf.impute_na(data,
                                el,
                                replacement=config.IMPUTATION_DICT[el])

    # Group rare labels
    for el in config.CATEGORICAL_VARS:
        data[el] = pf.remove_rare_labels(data, el, config.FREQUENT_LABELS[el])

    # encode variables
    for el in config.CATEGORICAL_VARS:
        data = pf.encode_categorical(data, el)

    # check all dummies were added
    data = pf.check_dummy_variables(data, config.DUMMY_VARIABLES)

    # scale variables
    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)

    return predictions

Example #12

def predict(data):

    # extract first letter from cabin
    data = pf.extract_cabin_letter(data, 'cabin')

    # impute NA categorical
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.impute_na(data, var, replacement='Missing')

    # impute NA numerical
    for var in config.NUMERICAL_TO_IMPUTE:
        data[var] = pf.impute_na(data,
                                 var,
                                 replacement=config.IMPUTATION_DICT[var])

    # Group rare labels
    for var in config.CATEGORICAL_VARS:
        data[var] = pf.remove_rare_labels(data, var,
                                          config.FREQUENT_LABELS[var])

    # encode variables
    for var in config.CATEGORICAL_VARS:
        data = pf.encode_categorical(data, var)

    # check all dummies were added
    for var in config.DUMMY_VARIABLES:
        if (var not in X_train.columns):
            data[var] = 0

    # scale variables
    data = pf.scale_features(data, config.OUTPUT_SCALER_PATH)

    # make predictions
    predictions = pf.predict(data, config.OUTPUT_MODEL_PATH)

    return predictions

Example #13

File: train.py Project: overdrivek/MLDeployment

import pandas as pd
import config
# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
df_data = pf.load_data(config.PATH_TO_DATASET)
df_target = df_data[config.TARGET]
df_data = df_data.drop([config.TARGET], axis=1)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(
    df_data, df_target, seed=config.GLOBAL_SEED)

# get first letter from cabin variable
X_train = pf.extract_cabin_letter(X_train,
                                  config.IMPUTATION_DICT['cabin_variable'])
X_test = pf.extract_cabin_letter(X_test,
                                 config.IMPUTATION_DICT['cabin_variable'])

# impute categorical variables
X_train = pf.add_missing_indicator(X_train, config.CATEGORICAL_VARS)
X_test = pf.add_missing_indicator(X_test, config.CATEGORICAL_VARS)

# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    X_train = pf.impute_na(X_train,
                           var,
                           replace_by=config.IMPUTATION_DICT[var],
                           add_na_columns=True)
    X_test = pf.impute_na(X_test,
                          var,

Example #14

import preprocessing_functions as pf
import config

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
data = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(data, config.TARGET)

# get first letter from cabin variable
X_train['cabin'] = pf.extract_cabin_letter(X_train, 'cabin')
X_test['cabin'] = pf.extract_cabin_letter(X_test, 'cabin')

# impute categorical variables
for var in config.CATEGORICAL_VARS:
    X_train[var + '_na'] = pf.add_missing_indicator(X_train, var)
    X_train[var] = pf.impute_na(X_train, var)

    X_test[var + '_na'] = pf.add_missing_indicator(X_test, var)
    X_test[var] = pf.impute_na(X_test, var)

# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    X_train[var + '_na'] = pf.add_missing_indicator(X_train, var)
    X_train[var] = pf.impute_na(X_train, var, config.IMPUTATION_DICT[var])
    X_test[var + '_na'] = pf.add_missing_indicator(X_test, var)
    X_test[var] = pf.impute_na(X_test, var, config.IMPUTATION_DICT[var])

Example #15

import preprocessing_functions as pf
import config

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
data = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(data, config.TARGET)


# get first letter from cabin variable
X_train['cabin'] = pf.extract_cabin_letter(X_train)


# impute categorical variables
for var in config.CATEGORICAL_VARS:
    X_train[var] = pf.impute_na(X_train, var)


# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    X_train = pf.add_missing_indicator(X_train, var)
    X_train[var] = pf.impute_na(X_train, var, config.IMPUTATION_DICT[var])



# Group rare labels
for var in config.CATEGORICAL_VARS:

Example #16

File: train.py Project: wonglimey/machine-learning-pipeline

import preprocessing_functions as pf
import config

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
data = pf.load_data(config.PATH_TO_DATASET)


# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(data, config.TARGET)


# get first letter from cabin variable
X_train[config.EXTRACT_VARIABLE] = pf.extract_cabin_letter(X_train, config.EXTRACT_VARIABLE)


# impute categorical variables
for var in config.CATEGORICAL_TO_ENCODE:
    X_train[var] = pf.impute_na(X_train, var, replacement='Missing')


# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    if (var == 'age'):
        X_train[var] = pf.add_missing_indicator(X_train, var, config.AGE_MEDIAN)
    else:
        X_train[var] = pf.add_missing_indicator(X_train, var, config.FARE_MEDIAN)

Example #17

import preprocessing_functions as pf
import config

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
df = pf.load_data(config.PATH_TO_DATASET)

# divide data set
xtrain, xtest, ytrain, ytest = pf.divide_train_test(df, config.TARGET)

# # get first letter from cabin variable
xtrain['cabin'] = pf.extract_cabin_letter(xtrain, 'cabin')

# # impute categorical variables
xtrain[config.CATEGORICAL_VARS] = pf.impute_na(xtrain[config.CATEGORICAL_VARS],
                                               'Missing')

# # impute numerical variable
xtrain[config.NUMERICAL_TO_IMPUTE] = pf.impute_na(
    xtrain[config.NUMERICAL_TO_IMPUTE], 'Numerical')

# # Group rare labels
for var in config.CATEGORICAL_VARS:
    xtrain[var] = pf.remove_rare_labels(xtrain, var,
                                        config.FREQUENT_LABELS[var])

# # encode categorical variables
xtrain = pf.encode_categorical(xtrain, config.CATEGORICAL_VARS)

Example #18

import config
import preprocessing_functions as pf

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
data = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(data, config.TARGET)

# get first letter from cabin variable
X_train.loc[:, "cabin"] = pf.extract_cabin_letter(X_train, "cabin")

# impute categorical variables
for var in config.CATEGORICAL_VARS:
    X_train.loc[:, var] = pf.impute_na(X_train, var, replacement="Missing")

# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    median_val = X_train[var].median()
    X_train.loc[:, var] = pf.impute_na(X_train, var, replacement=median_val)

# Group rare labels
for var in config.CATEGORICAL_VARS:
    X_train.loc[:, var] = pf.remove_rare_labels(X_train, var,
                                                config.FREQUENT_LABELS[var])

# encode categorical variables
for var in config.CATEGORICAL_VARS:

Example #19

File: train.py Project: Nachimak28/ml_dep_assignments

import preprocessing_functions as pf
import config

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
data = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(data, config.TARGET)

# get first letter from cabin variable
data = pf.extract_cabin_letter(data, 'cabin')

# impute categorical variables
for var in config.CATEGORICAL_VARS:
    X_train[var] = pf.impute_na(X_train, var, replacement='Missing')

# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    X_train[var] = pf.impute_na(X_train,
                                var,
                                replacement=config.IMPUTATION_DICT[var])

# Group rare labels
for var in config.CATEGORICAL_VARS:
    X_train[var] = pf.remove_rare_labels(X_train, var,
                                         config.FREQUENT_LABELS[var])

# encode categorical variables

Example #20

import preprocessing_functions as pf
import config

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
df = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(df, config.TARGET)

# get first letter from cabin variable
pf.extract_cabin_letter(X_train, 'cabin')

# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    pf.add_missing_indicator(X_train, var)
    pf.impute_na(X_train, var, config.IMPUTATION_DICT[var])

# impute categorical variables
for var in config.CATEGORICAL_VARS:
    pf.impute_na(X_train, var)

# Group rare labels
for var, labels in config.FREQUENT_LABELS.items():
    pf.remove_rare_labels(X_train, var, labels)

# encode categorical variables
for var in config.CATEGORICAL_VARS:
    X_train = pf.encode_categorical(X_train, var)

Example #21

import config

import warnings
warnings.simplefilter(action='ignore')

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
data = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, _, y_train, _ = pf.divide_train_test(data, config.TARGET)

# get first letter from cabin variable
X_train["cabin"] = pf.extract_cabin_letter(X_train, "cabin")

# impute categorical variables
for var in config.CATEGORICAL_VARS:
    X_train[var] = pf.impute_na(X_train, var)

# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    X_train = pf.add_missing_indicator(X_train, var)
    X_train[var] = pf.impute_na(X_train, var, config.IMPUTATION_DICT[var])

# Group rare labels
for var in config.CATEGORICAL_VARS:
    X_train[var] = pf.remove_rare_labels(X_train, var,
                                         config.FREQUENT_LABELS[var])

Example #22

File: train.py Project: rmania/ml_deployment

import preprocessing_functions as pf
import config

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
data = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(data, config.TARGET)

# get first letter from cabin variable
X_train['cabin'] = pf.extract_cabin_letter(df=X_train, var='cabin')
X_test['cabin'] = pf.extract_cabin_letter(df=X_test, var='cabin')

# impute categorical variables
for var in config.CATEGORICAL_VARS:
    X_train[var] = pf.impute_na(X_train, var, replacement='Missing')
    X_test[var] = pf.impute_na(X_test, var, replacement='Missing')

# impute numerical variable
for var in config.NUMERICAL_TO_IMPUTE:
    X_train[var] = pf.add_missing_indicator(df=X_train, var=var)
    X_test[var] = pf.add_missing_indicator(df=X_test, var=var)

# Group rare labels
for var in config.CATEGORICAL_VARS:
    X_train[var] = pf.remove_rare_labels(X_train, var,
                                         config.FREQUENT_LABELS[var])
    X_test[var] = pf.remove_rare_labels(X_test, var,

Example #23

import preprocessing_functions as pf
import config
import pandas as pd

# ================================================
# TRAINING STEP - IMPORTANT TO PERPETUATE THE MODEL

# Load data
df = pf.load_data(config.PATH_TO_DATASET)

# divide data set
X_train, X_test, y_train, y_test = pf.divide_train_test(df, config.TARGET)

# get first letter from cabin variable
X_train = pf.extract_cabin_letter(X_train, 'cabin')
X_test = pf.extract_cabin_letter(X_test, 'cabin')

# impute categorical variables
X_train = pf.impute_na(X_train, config.CATEGORICAL_VARS)
X_test = pf.impute_na(X_test, config.CATEGORICAL_VARS)

# impute numerical variable
for var in config.IMPUTATION_DICT.keys():
    X_train = pf.add_missing_indicator(X_train, var)
    X_test = pf.add_missing_indicator(X_test, var)

    X_train = pf.impute_na(X_train, var, config.IMPUTATION_DICT[var])
    X_test = pf.impute_na(X_test, var, config.IMPUTATION_DICT[var])

# Group rare labels
for var in config.FREQUENT_LABELS.keys():