Using Libraries

import pandas as pd
import numpy as np
from pandas import Series, DataFrame
import scipy
from scipy import stats
import sklearn
from sklearn import preprocessing
from sklearn.preprocessing import scale
import seaborn as sb
from sklearn.model_selection import KFold
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score


#OPEN CSV
filename = 'C:/Users/n0762538/Documents/Data/WheatSeedsDatasetPractice/seeds_dataset2.csv'
datasetm = pd.read_csv(filename)
datasetm.columns =['1', '2', '3', '4', '5', '6', '7', 'outcome']
datasetm.head()


#PREPROCESS DATA

#Check for missing value
datasetm.isnull().sum()

#Get max and min values in each column
max = datasetm.max()
min = datasetm.min()

#Scaling inputs
inputs_1 = datasetm.iloc[0:209,0:7]
y = datasetm.iloc[0:209, 7:8]
scaled = preprocessing.MinMaxScaler()
scaled_dataset = scaled.fit_transform(inputs_1)
inputs = pd.DataFrame(scaled_dataset)

#New dataset
outputs = datasetm.iloc[0:209,-1]
output = pd.Series(outputs)
dataset = pd.concat([inputs,output], axis=1)
print (dataset)

#Split into folds, train and test set
kf = KFold(n_splits=7)
for train_index, test_index in kf.split(dataset):
    print("TRAIN:", train_index, "TEST:", test_index)

#Accuracy
# Calculate accuracy percentage
def accuracy_metric(actual, predicted):
    return accuracy_score(actual, predicted)