import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score, confusion_matrix
from sklearn import metrics
from sklearn.model_selection import train_test_split


dataframe = pd.read_csv("archive/data.csv")

# Change bool values to int (0/1)
dataframe = dataframe.rename(columns={'Winner': 'Blue_Corner_Win'})
dataframe['Blue_Corner_Win'] = (dataframe['Blue_Corner_Win'] == 'Blue').astype(int)
dataframe['title_bout'] = dataframe['title_bout'].astype(int)

# Select datas for Yi (blue corner win ?) and columns for stat
Yi = dataframe[["Blue_Corner_Win"]]
colonnes = ['B_avg_BODY_landed', 'B_avg_HEAD_landed', 'B_avg_TD_att', 'B_avg_TOTAL_STR_landed', 
                   'B_avg_opp_BODY_att', 'B_avg_opp_HEAD_landed', 'B_avg_opp_LEG_landed', 
                   'B_avg_opp_SIG_STR_att', 'B_avg_opp_TOTAL_STR_att', 'R_avg_TD_att', 'R_avg_opp_GROUND_att', 
                   'R_avg_opp_SIG_STR_landed', 'B_age', 'R_age']
                   
Xi = dataframe[colonnes]

Xtrain, Xtest, ytrain, ytest = train_test_split(Xi, Yi,test_size=0.20, random_state=42)
print(Xtrain.shape)
print(Xtest.shape)

Arbre_decision = DecisionTreeClassifier(random_state=0, max_depth=20)
clf = Arbre_decision.fit(Xtrain, Yi)

ypredict = clf.predict(Xtest)
accuracy = accuracy_score(ytest, ypredict)
matriceConfusion = confusion_matrix(ytest, ypredict)

incorrect=matriceConfusion[0][1] + matriceConfusion[1][0]
total = matriceConfusion.sum()

print("\nNumber of incorrect classifications: " + str(incorrect))
print("Number of classifications total: " + str(total))

print("Percent: "+ str((total-incorrect)/total*100))

fighter_data = dataframe[dataframe['R_fighter'] == 'Adrian Yanez']
average_fighter_data = fighter_data[colonnes].mean()

fighter_data_2 = dataframe[dataframe['R_fighter'] == 'Gustavo Lopez']
average_fighter_data_2 = fighter_data_2[colonnes].mean()

combined_features = pd.concat([average_fighter_data, average_fighter_data_2])

prediction = clf.predict([combined_features])
if prediction[0] == 1:
    winner = "Blue Corner"
else:
    winner = "Red Corner"

print(f"The predicted winner is: {winner}")