topiis.py

import pandas as pd
import os
import sys


def main():
    # Argument check: args = 5
    if len(sys.argv) != 5:
        args_check = "greater" if (len(sys.argv) > 5) else "less"
        print("Error: Number of arguments is ", args_check, " than 5")
        print("Usage: python topsis.py <InputDataFile> <Weights> <Impacts> <ResultFileName>")
        exit(1)

    # Input Data File check
    elif not os.path.isfile(sys.argv[1]):
        print(f"ERROR: Input data file doesn't exists! : {sys.argv[1]}")
        exit(1)

    # File extension check
    elif ".csv" != (os.path.splitext(sys.argv[1]))[1]:
        print(f"ERROR: Input data file is not in csv format! : {sys.argv[1]}")
        exit(1)

    else:
        dataset, temp_dataset = pd.read_csv(
            sys.argv[1]), pd.read_csv(sys.argv[1])
        nCol = len(temp_dataset.columns.values)

        # less than 3 columns in input dataset
        if nCol < 3:
            print("ERROR: Input data file have less than 3 columns")
            exit(1)

        # Handling non-numeric value
        for i in range(1, nCol):
            pd.to_numeric(dataset.iloc[:, i], errors='coerce')
            dataset.iloc[:, i].fillna(
                (dataset.iloc[:, i].mean()), inplace=True)

        # Handling input weights and impacts args
        try:
            weights = [int(i) for i in sys.argv[2].split(',')]
        except:
            print("ERROR: Either the weights contains non-integers or not comma sperated")
            exit(1)

        impact = sys.argv[3].split(',')
        for i in impact:
            if not (i == '+' or i == '-'):
                print(
                    "ERROR: Either the impacts contains symbol other than '+' and '-' or are not correctly comma "
                    "seperated")
                exit(1)

        # Checking number of column,weights and impacts is same or not
        if nCol != len(weights) + 1 or nCol != len(impact) + 1:
            print(
                "ERROR : Number of weights, number of impacts and number of columns not same")
            exit(1)

        if ".csv" != (os.path.splitext(sys.argv[4]))[1]:
            print("ERROR : Output file extension is wrong")
            exit(1)
        if os.path.isfile(sys.argv[4]):
            os.remove(sys.argv[4])
        # print(" No error found\n\n Applying Topsis Algorithm...\n")
        topsis_pipy(temp_dataset, dataset, nCol, weights, impact)


def normalize(temp_dataset, nCol, weights):
    for i in range(1, nCol):
        temp = 0
        for j in range(len(temp_dataset)):
            temp = temp + temp_dataset.iloc[j, i] ** 2
        temp = temp ** 0.5

        for j in range(len(temp_dataset)):
            temp_dataset.iat[j, i] = (
                                             temp_dataset.iloc[j, i] / temp) * weights[i - 1]
    return temp_dataset


def calc_values(temp_dataset, nCol, impact):
    p_sln = temp_dataset.max().values[1:]
    n_sln = temp_dataset.min().values[1:]
    for i in range(1, nCol):
        if impact[i - 1] == '-':
            p_sln[i - 1], n_sln[i - 1] = n_sln[i - 1], p_sln[i - 1]
    return p_sln, n_sln


def topsis_pipy(temp_dataset, dataset, nCol, weights, impact):
    temp_dataset = normalize(temp_dataset, nCol, weights)
    p_sln, n_sln = calc_values(temp_dataset, nCol, impact)

    score = []
    for i in range(len(temp_dataset)):
        temp_p, temp_n = 0, 0
        for j in range(1, nCol):
            temp_p += (p_sln[j - 1] - temp_dataset.iloc[i, j]) ** 2
            temp_n += (n_sln[j - 1] - temp_dataset.iloc[i, j]) ** 2
        temp_p, temp_n = temp_p ** 0.5, temp_n ** 0.5
        score.append(temp_n / (temp_p + temp_n))
    dataset['Topsis Score'] = score

    dataset['Rank'] = (dataset['Topsis Score'].rank(
        method='max', ascending=False))
    dataset = dataset.astype({"Rank": int})

    dataset.to_csv("result.csv", index=False)
    print(dataset)