!pip uninstall scikit-learn # until no more scikit-learn is present
!pip install scikit-learn
!pip install scikit-optimize
!pip install skll
!pip install imbalanced-learn
!pip install eli5
!pip install scipy
!pip install scikit-optimize

# activate R magic to run R in google colab notebook
import rpy2
%load_ext rpy2.ipython

#%%R

#install.packages("MLmetrics")
#install.packages("yardstick")
#install.packages("mltools")
#install.packages("glue")

%tensorflow_version 2.x
import numpy as np
import pandas as pd
import io
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns; sns.set(style="ticks", color_codes=True)
from sklearn import preprocessing
from sklearn.preprocessing import StandardScaler, RobustScaler
from sklearn.preprocessing import MinMaxScaler
from sklearn.feature_selection import SelectKBest, f_classif
from sklearn.feature_selection import chi2
from sklearn.linear_model import LogisticRegression
from sklearn import feature_selection
#from sklearn.preprocessing import Imputer
from sklearn.model_selection import  cross_val_score
from sklearn.pipeline import Pipeline
from sklearn.metrics import make_scorer
from sklearn.preprocessing import LabelEncoder
from sklearn.metrics import accuracy_score
from sklearn.model_selection import cross_validate
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import make_classification
from sklearn.model_selection import StratifiedKFold
from sklearn.feature_selection import RFECV
from xgboost import XGBClassifier
from sklearn.impute import SimpleImputer
import xgboost as xgb
import lightgbm as lgb
from sklearn.ensemble import ExtraTreesClassifier
from sklearn.linear_model import LogisticRegression
from sklearn import svm
from skopt.space import Real, Categorical, Integer
#from skll.metrics import spearman
from scipy.stats import kendalltau, spearmanr, pearsonr
from skopt import BayesSearchCV
from sklearn.model_selection import  cross_val_score
from sklearn.pipeline import Pipeline
from sklearn.metrics import make_scorer
from skopt.space import Real, Categorical, Integer
from sklearn.metrics import classification_report
from sklearn.base import TransformerMixin
from sklearn.metrics import classification_report, f1_score, accuracy_score, precision_score, confusion_matrix
from sklearn.metrics import roc_auc_score
from sklearn.metrics import precision_recall_curve
pd.set_option('display.max_rows', 600)
pd.set_option('display.max_columns', 500)
pd.set_option('display.width', 1000)
import warnings
import pandas_profiling
from sklearn.ensemble import ExtraTreesClassifier
from sklearn.feature_selection import SelectFromModel
from sklearn.base import BaseEstimator, TransformerMixin
from sklearn.ensemble import ExtraTreesClassifier
from sklearn.pipeline import make_pipeline, FeatureUnion, Pipeline
from sklearn.preprocessing import OneHotEncoder,  StandardScaler
from sklearn.model_selection import GridSearchCV
from imblearn.over_sampling import SMOTE
from sklearn.model_selection import train_test_split, RandomizedSearchCV
from sklearn.metrics import roc_curve, roc_auc_score,balanced_accuracy_score
from sklearn.svm import SVC
import random
import matplotlib.pyplot as plt
import seaborn as sns
import re
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.compose import ColumnTransformer, make_column_transformer
from imblearn.over_sampling import RandomOverSampler
from imblearn.under_sampling import RandomUnderSampler
from sklearn.ensemble import AdaBoostClassifier
from sklearn.ensemble import *
from sklearn.utils import resample
from imblearn.over_sampling import SMOTE
#import smote
import os
from sklearn.tree import DecisionTreeClassifier
from imblearn.metrics import geometric_mean_score as gmean
from imblearn.metrics import make_index_balanced_accuracy as iba
from imblearn.metrics import *
from eli5.sklearn import PermutationImportance
from eli5.sklearn import *
import eli5
from eli5.permutation_importance import get_score_importances
#import rus
# Skopt functions
from skopt import BayesSearchCV
from skopt import gp_minimize # Bayesian optimization using Gaussian Processes
from skopt.space import Real, Categorical, Integer
from skopt.utils import use_named_args # decorator to convert a list of parameters to named arguments
from skopt.callbacks import DeadlineStopper # Stop the optimization before running out of a fixed budget of time.
from skopt.callbacks import VerboseCallback # Callback to control the verbosity
from skopt.callbacks import DeltaXStopper # Stop the optimization If the last two positions at which the objective has been evaluated are less than delta
from joblib import dump, load
from prettytable import PrettyTable
from collections import Counter
from sklearn.datasets import make_classification
from imblearn.under_sampling import RandomUnderSampler # doctest: +NORMALIZE_WHITESPACE
from imblearn import under_sampling, over_sampling
from imblearn.over_sampling import SMOTE
import tensorflow as tf
warnings.filterwarnings("ignore")
%matplotlib inline
#specify tensorflow version to use
%tensorflow_version 2.x
#load tensorboard
#%load_ext tensorboard
 #%tensorboard --logdir logs

%autosave 5

np.version.version

from sklearn.impute import SimpleImputer
imputer = SimpleImputer(missing_values=np.nan, strategy='mean')

file = tf.keras.utils
df = pd.read_csv('https://storage.googleapis.com/download.tensorflow.org/data/creditcard.csv')
df.head()

df[['Time', 'V1', 'V2', 'V3', 'V4', 'V5', 'V26', 'V27', 'V28', 'Amount', 'Class']].describe().transpose()

df['Class'].value_counts(normalize=True)*100

neg, pos = df.Class.value_counts()
total = neg + pos
print('Examples:\n    Total: {}\n    Positive: {} ({:.2f}% of total)\n  '.format(
    total, pos, 100 * pos / total,100 * neg / total))

print('Total: {}\n    Negative: {} ({:.2f}% of total)\n  '.format(
    total, neg, 100 * neg / total))

#x = raw_df.drop(['Time'],axis=1)

# Use a utility from sklearn to split and shuffle our dataset.
train_df, test_df = train_test_split(df, test_size=0.2)
#train_df, val_df = train_test_split(train_df, test_size=0.2)

train_x =train_df.drop(['Time','Class'],axis=1)
test_x = test_df.drop(['Time','Class'],axis=1)
#val_x  =  val_df.drop(['Time'],axis=1)

train_y=  train_df.Class
test_y = test_df.Class
#val_y  = val_df.Class

print('Traing dataset size:{}'.format(train_x.shape))
print('Test dataset size:{}'.format(test_x.shape))
#print('Validation dataset size: {}'.format(val_df.shape))

#train_x.columns
#test_x.columns
test_y.isna().sum()

# Setting a 5-fold stratified cross-validation (note: shuffle=True)
skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=0)


clf = xgb.XGBClassifier(
        n_jobs = -1,
        objective = 'binary:logistic',
        silent=1,
        tree_method='approx')

search_spaces = {
    #'learning_rate': Real(0.01, 1.0, 'log-uniform'),
    #             'min_child_weight': Integer(0, 10),
    #             'max_depth': Integer(1, 50),
    #            'max_delta_step': Integer(0, 20), # Maximum delta step we allow each leaf output
    #             'subsample': Real(0.01, 1.0, 'uniform'),
    #             'colsample_bytree': Real(0.01, 1.0, 'uniform'), # subsample ratio of columns by tree
    #             'colsample_bylevel': Real(0.01, 1.0, 'uniform'), # subsample ratio by level in trees
                 #'reg_lambda': Real(1e-9, 1000, 'log-uniform'), # L2 regularization
                 #'reg_alpha': Real(1e-9, 1.0, 'log-uniform'), # L1 regularization
     #            'gamma': Real(1e-9, 0.5, 'log-uniform'), # Minimum loss reduction for partition/pruning parameter
     #            'n_estimators': Integer(50, 100),
                 'scale_pos_weight': Real(1e-6, 2000, 'log-uniform')
                 }


bayessearch = BayesSearchCV(clf,
                    search_spaces,
                    scoring='roc_auc', #f1
                    cv=skf,
                    n_iter=40,
                    n_jobs=-1,
                    return_train_score=False,
                    #refit=True,
                    optimizer_kwargs={'base_estimator': 'GP'},
                    random_state=22)
    

#xgbm_model = bayessearch.fit(X=train_x,  y=train_y)

from google.colab import drive
drive.mount('/content/drive')

#share_link="https://drive.google.com/file/d/1mGzO4-vaTKVgH5zzzXCVcXNbj-Bt8u8g/view?usp=sharing"
import os 
os.getcwd()
#!files.os.listdir()

#Inline report without saving object
pandas_profiling.ProfileReport(df)
#Save report to file¶
pfr = pandas_profiling.ProfileReport(df)

pfr.to_file("/content/drive/My Drive/profilingReport2.html")
pfr