import pandas as pd
import numpy as np
import os

from sklearn.datasets import load_iris
from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import f_classif
from sklearn.model_selection import RepeatedStratifiedKFold
from sklearn.neural_network import MLPClassifier

features_dic = {}
results_dic = {}

script_dir = os.path.dirname(__file__)
rel_path = "bialaczka_switched.xls"
abs_file_path = os.path.join(script_dir, rel_path)

dataExcel = pd.read_excel(abs_file_path, nrows=410)
df = pd.DataFrame(dataExcel)

feature_data = df.iloc[:, :-2]
diagnose_classes = np.array(df['Klasa'])

# Create an SelectKBest object to select features with two best ANOVA F-Values
fvalue_selector = SelectKBest(f_classif)

# Apply the SelectKBest object to the features and target
feature_data_kbest = fvalue_selector.fit(feature_data, diagnose_classes)

ranking = fvalue_selector.scores_

ranking_ = pd.DataFrame(ranking)

X = fvalue_selector.transform(feature_data)
y = diagnose_classes


rskf = RepeatedStratifiedKFold(n_splits=2, n_repeats=2,
    random_state= 36851234)
for train_index, test_index in rskf.split(X, y):
    # print("TRAIN:", train_index, "TEST:", test_index)
    X_train, X_test = X[train_index], X[test_index]
    y_train, y_test = y[train_index], y[test_index]

clf = MLPClassifier(hidden_layer_sizes=(100), max_iter=600, alpha=0.0001,
                    solver='sgd',learning_rate='constant',momentum =0,
                    learning_rate_init=0.2, verbose=10,  random_state=21,tol=0.000000001)

clf.fit(X_train, y_train)
y_pred = clf.predict(X_test)

print(y_pred)