SKLearn pipeline

gistfile1.txt

import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
from sklearn.metrics import roc_auc_score
from sklearn.preprocessing import LabelEncoder, StandardScaler 
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import make_pipeline
from sklearn.pipeline import Pipeline
from lightgbm import LGBMClassifier
from xgboost import XGBClassifier
from category_encoders import OneHotEncoder
from sklearn.model_selection import cross_val_predict
from warnings import filterwarnings
from sklearn.impute import SimpleImputer


#Import data
train = pd.read_csv("../input/titanic/train.csv")
print("train shape", train.shape)
test = pd.read_csv("../input/titanic/test.csv")
print("test shape", test.shape)


#Target
target_column = "Survived"
id_column = "PassengerId"
categorical_cols = [c for c in test.columns if test[c].dtype in [np.object]]
numerical_cols = [c for c in test.columns if test[c].dtype in [np.float, np.int] and c not in [target_column, id_column]]
print("Number of features", len(categorical_cols)+len(numerical_cols))



#Classifer

numeric_transformer = Pipeline(steps=[('imputer', SimpleImputer(strategy='median')),('scaler', StandardScaler())])
categorical_transformer = Pipeline(steps=[('imputer', SimpleImputer(strategy='constant', fill_value='missing')),('onehot', OneHotEncoder(handle_unknown='ignore'))])

pipe = make_pipeline(
    ColumnTransformer([
        ('num', numeric_transformer, numerical_cols),
        ('cat', categorical_transformer, ['Embarked','Sex']),    
    ]),
    RandomForestClassifier(),
)

prediction = cross_val_predict(pipe, train, train[target_column], cv=5, method="predict_proba")
print("Cross validation AUC {:.4f}".format(roc_auc_score(train[target_column], prediction[:,1])))



from sklearn.metrics import accuracy_score, log_loss
from sklearn.neighbors import KNeighborsClassifier
from sklearn.svm import SVC, LinearSVC, NuSVC
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier, GradientBoostingClassifier
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis
classifiers = [
    RandomForestClassifier(),    
    KNeighborsClassifier(3),
    SVC(kernel="rbf", C=0.025, probability=True),
    NuSVC(probability=True),
    DecisionTreeClassifier(),
    RandomForestClassifier(),
    AdaBoostClassifier(),
    GradientBoostingClassifier(),
    XGBClassifier(),
]
for classifier in classifiers:
    pipe = make_pipeline(
    ColumnTransformer([
        ('num', numeric_transformer, numerical_cols),
        ('cat', categorical_transformer, ['Embarked','Sex']),    
    ]),
    classifier
    )
    prediction = cross_val_predict(pipe, train, train[target_column], cv=5, method="predict_proba")
    print(classifier)
    print("Cross validation AUC {:.4f}".format(roc_auc_score(train[target_column], prediction[:,1])))
    
    
 rf = Pipeline(steps=[('preprocessor',  ColumnTransformer([
        ('num', numeric_transformer, numerical_cols),
        ('cat', categorical_transformer, ['Embarked','Sex']),    
    ]),
    ),
    ('classifier', RandomForestClassifier())])

param_grid = { 
    'classifier__n_estimators': [200, 300],
    'classifier__max_features': ['auto', 'sqrt', 'log2'],
    'classifier__max_depth' : [4,6,8],
    'classifier__criterion' :['gini', 'entropy']
}
from sklearn.model_selection import GridSearchCV
CV = GridSearchCV(rf, param_grid, cv = 3, n_jobs = -1, verbose = 1)      
CV.fit(train, train[target_column])  
print(CV.best_params_)    
print(CV.best_score_)


import eli5
pipe.fit(train, train[target_column])
eli5.show_weights(pipe, top=10)