PyPI 官网下载 | ucs633_Aditri-0.1.tar.gz

# -*- coding: utf-8 -*- """ Created on Sun Jan 17 19:41:38 2020 @author: Aditri """ import numpy as np import sys from sklearn.preprocessing import LabelEncoder import pandas as pd from scipy.stats import rankdata from tabulate import tabulate # if not installed already enter command ->pip install tabulate def main(): # Checking proper inputs on command prompt if len(sys.argv)!=4: print("Incorrect parameters.Input format:python <programName> <dataset> <weights array> <impacts array>") exit(1) else: # Importing dataset data = pd.read_csv(sys.argv[1]).values # Dropping the first column which is serves as the index in the sample dataset data1 = data[:,1:] # Initialising the weights array w = [int(i) for i in sys.argv[2].split(',')] # Initialising the impact array imp = sys.argv[3].split(',') # y is the last column which contains categorical values in string format in the sample dataset y=data1[:,-1] # Encoding the categorical values and adding it back to the dataset label_encoder = LabelEncoder() y = label_encoder.fit_transform(y) data1[:,-1]=y topsis(data1 , w , imp) def topsis(data1,w,imp): r = data1.shape[0] c = data1.shape[1] if len(w) != c or len(imp) != c or not all(i > 0 for i in w) or not all(i=="+"or i=="-" for i in imp): return print("Please check the input arguments.") data = np.zeros([r+2,c+4]) s=sum(w) for i in range(c): for j in range(r): data[j,i] = (data1[j,i]/np.sqrt(sum(data1[:,i]**2)))*w[i]/s for i in range(c): data[r,i] = max(data[:r,i]) data[r+1,i] = min(data[:r,i]) if i[i] == "-": data[r,i] , data[r+1,i] = data[r+1,i] , data[r,i] for i in range(r): data[i,c] = np.sqrt(sum((data[r,:c] - data[i,:c])**2)) data[i,c+1] = np.sqrt(sum((data[r+1,:c] - data[i,:c])**2)) data[i,c+2] = data[i,c+1]/(data[i,c] + data[i,c+1]) data[:r,c+3] = len(data[:r,c+2]) - rankdata(data[:r,c+2]).astype(int) + 1 print(tabulate({"Model": np.arange(1,r+1), "Performance Score": data[:5,c+2], "Rank": data[:5,c+3]}, headers="keys")) if __name__ == "__main__": main()