import pandas as pd
from sklearn.tree import  DecisionTreeClassifier, export_graphviz
from sklearn.model_selection import train_test_split
import subprocess

import sklearn.metrics as met


def visualize_tree(tree, feature_names, class_names):

    with open("dt.dot", "w") as f:
        export_graphviz(tree, out_file=f, feature_names=feature_names,
                        class_names=class_names,
                        filled=True, rounded=True)

    f.close()

    subprocess.call("dot -Tpng dt.dot -o tree.png", shell=True)


#ucitavanje podataka
df = pd.read_csv("iris.csv")


#prikaz imena kolona + 5 prvih instanci
print('Prvih 5 instanci', df.head(), sep='\n')
print('\n\n')


print('Opis podataka', df.describe(), sep='\n')
print('\n\n')

print('Korelacija', df.corr(), sep='\n')
print('\n\n')


#prikaz imena kolona + 5 poslednjih instanci
#print(df.tail())
#print('\n\n')


#prikaz klasa
print("Klase: ", df["Species"].unique())
print('\n\n')

featurs = df.columns[:4].tolist()
print('Atributi za pravljenje modela:', featurs)
print('\n\n')

x=df[featurs]
y=df["Species"]

x_train, x_test, y_train, y_test = train_test_split(x, y, train_size=100, stratify=y)
#dt = DecisionTreeClassifier()
#dt = DecisionTreeClassifier(criterion='entropy')
#dt = DecisionTreeClassifier(max_depth=2)
#dt = DecisionTreeClassifier(min_samples_split=20, max_depth=4)
dt = DecisionTreeClassifier(min_samples_split=20, max_depth=4, max_leaf_nodes=4)
#dt = DecisionTreeClassifier(min_impurity_split=0.05)

#pravljenje drveta odlucivanja na osnovu trening skupa
dt.fit(x_train, y_train)


print('Predvidjena verovatnoca', dt.predict_proba(x_test), sep='\n')
print('\n\n')

#graficki prikaz drveta
visualize_tree(dt, featurs, df['Species'].unique())

#primena modela na test podacima
y_pred = dt.predict(x_test)

cnf_matrix = met.confusion_matrix(y_test, y_pred)
print('Matrica konfuzije', cnf_matrix, sep='\n')
print('\n\n')

accuracy = met.accuracy_score(y_test, y_pred)
print('Preciznost', accuracy)

accuracy = met.accuracy_score(y_test, y_pred, normalize=False)
print('Preciznost u broju instanci', accuracy, )
print('\n\n')

class_report = met.classification_report(y_test, y_pred, target_names=df['Species'].unique())
print('Izvestaj klasifikacije', class_report, sep='\n')