import pandas as pd
import  matplotlib.pyplot  as  plt
from sklearn import svm
import numpy as np
from  sklearn.preprocessing  import  StandardScaler
from  sklearn.neural_network  import  MLPClassifier
from  sklearn.metrics  import classification_report,confusion_matrix

data =pd.read_csv( 'titanic_800.csv' ,  sep = ',' , header = 0,
                  usecols=['Survived','Sex','Pclass','Age','SibSp','Parch','Fare','Embarked'])
yvalues = pd.DataFrame( dict (Survived =[]),  dtype = int )

yvalues["Survived"] =  data["Survived"].copy()
X = data.drop( 'Survived' ,  axis = 1 ,  inplace = True )


data["Embarked"] = data["Embarked"].replace('C',0)
data["Embarked"] = data["Embarked"].replace('S',1)
data["Embarked"] = data["Embarked"].replace('Q',2)

data["Sex"] = data["Sex"].replace('male',0)
data["Sex"] = data["Sex"].replace('female',1)

x = data["Age"]
y = data["Pclass"]
a = data["SibSp"]
b = data["Parch"]
c = data["Fare"]
d = data["Embarked"]
e = data["Sex"]



plt.figure()
plt.scatter(x.values, yvalues.values,  color = 'black' ,  s = 30 )
plt.show()

plt.figure()
plt.scatter(y.values, yvalues.values,  color = 'green' ,  s = 30 )
plt.show()

plt.figure()
plt.scatter(a.values, yvalues.values,  color = 'red' ,  s = 30 )
plt.show()

plt.figure()
plt.scatter(b.values, yvalues.values,  color = 'black' ,  s = 30 )
plt.show()

plt.figure()
plt.scatter(c.values, yvalues.values,  color = 'green' ,  s = 30 )
plt.show()

plt.figure()
plt.scatter(d.values, yvalues.values,  color = 'red' ,  s = 30 )
plt.show()

plt.figure()
plt.scatter(e.values, yvalues.values,  color = 'black' ,  s = 30 )
plt.show()



newdata = data.fillna(0.0)
print(str(newdata))
xtrain  = newdata.head(700)
ytrain = yvalues.head(700)
xtest = newdata.tail(100)
ytest = yvalues.tail(100)


scaler = StandardScaler().fit(xtrain)
xtrain = scaler.transform(xtrain)
xtest= scaler.transform(xtest)

mlp = MLPClassifier(hidden_layer_sizes= (20,17,13),max_iter = 1000,random_state = 0, alpha = 0.1)
mlp.fit(xtrain,ytrain.values.ravel())
predictions = mlp.predict(xtest)

matrix = confusion_matrix(ytest,predictions)
print ('Confusion Matrix for MLP classifier')
print (matrix)
print()

print()
print('Classification report for MLP classifier')
print(classification_report(ytest,predictions))


clf = svm.SVC(C=1.0, kernel='poly', gamma = 0.1, random_state=None)
clf.fit(xtrain, np.ravel(ytrain))

predictions1 = clf.predict(xtest)
matrix1 = confusion_matrix(ytest,predictions1)
print ('Confusion Matrix for SVM')
print(matrix1)
print()
print('Classification report for SVM')
print(classification_report(ytest,predictions1))