Diabetic Classification: A Rigorous Analysis

from google.colab import drive
drive.mount('/content/gdrive')

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import accuracy_score
from sklearn.metrics import zero_one_loss
from sklearn.preprocessing import StandardScaler

def load_data():

#Load Data
  dataset = pd.read_csv('/content/gdrive/MyDrive/Robotech/ML/week_2/diabetes.csv')

#The values of some columns are not measured.
  zero_not_accepted = ['Glucose', 'BloodPressure', 'SkinThickness', 'BMI', 'Insulin']

  for column in zero_not_accepted:

    #Set the data whose value is zero to 'Nan'
    dataset[column] = dataset[column].replace(0, np.nan)

    #Calculate the average regardless of Nan values
    mean = int(dataset[column].mean(skipna=True))

    #Substituting the average value for the Nan values
    dataset[column] = dataset[column].replace(np.nan, mean)

#Separation of 'Data(features)' from dataset
  data = dataset.iloc[:, 0:8]

#Separation of 'Label' from dataset
  label = dataset.iloc[:, 8]

  #Separation of 'Test Data' and 'Train Data' from dataset
  #20% for test data
  #80% for train data

  '''
  x_train: Train Features
  x_test: Test Features
  y_train: Train Label
  y_test: Test Label
  '''
  x_train, x_test, y_train, y_test = train_test_split(data, label, test_size=0.2, random_state=42)

  return x_train, x_test, y_train, y_test

  def algorithm():
  #Training Model
  clf = KNeighborsClassifier(n_neighbors=11, metric= "euclidean")
  clf.fit(x_train, y_train)

return clf

def show_result():

  y_pred = clf.predict(x_test)

  #Accuracy
  acc = accuracy_score(y_test, y_pred)
  print("Accuracy: {:.2f}".format(acc*100))

  #Loss
  loss = zero_one_loss(y_test, y_pred)
  print("Loss: {:.2f}".format(loss*100))

x_train, x_test, y_train, y_test = load_data()

clf = algorithm()

show_result()