from __future__ import print_function

import pandas as pd

import numpy as np

import seaborn as sns

import matplotlib.pyplot as plt

from sklearn.neighbors import KNeighborsClassifier

from sklearn.preprocessing import LabelBinarizer

import warnings

from sklearn.preprocessing import MinMaxScaler

warnings.filterwarnings('ignore', module='sklearn')

### IMPORT DATASET

data = pd.read_csv('C:/Users/Student/Desktop/PROJECT/DataSet.CSV',header='infer')

### VISUALIZATION

print("\n\nHERE IS THE HEAD")

print(data.head())

print("\n\nHERE ARE THE COLUMNS")

print(data.columns)

print("\n\nHERE IS THE DESCRIPTION")

print(data.describe())

### PLOTTING GRAPH

# ax = plt.axes()

# ax.scatter(data.Age, data.Snoring)

# ax.set(xlabel='Age',

# ylabel='Snoring',

# title='Table');

# print(plt.show())

### CONVERTING TO FLOAT

lb = LabelBinarizer()

for col in ['Patient Id' ,'Level']:

data[col] = lb.fit_transform(data[col])

msc = MinMaxScaler()

data = pd.DataFrame(msc.fit_transform(data),

columns=data.columns)

### DATAFRAMES

x_cols = [x for x in data.columns if x != 'Level']

X_data = data[x_cols]

y_data = data['Level']

### KNN PREDICTION

knn = KNeighborsClassifier(n_neighbors=3)

knn = knn.fit(X_data,y_data)

y_pred = knn.predict(X_data)

print(y_pred)

### ACCURACY

def accuracy(real, predict):

return sum(y_data == y_pred) / float(real.shape[0])

print(accuracy(y_data, y_pred))