Python库 | Topsis_101903244-0.1.tar.gz

import pandas as pd import sys import numpy as np import logging def topsis_score(fname,w,im,outfile): logging.basicConfig(filename="101903244-log.log",level = logging.DEBUG,encoding='utf-8') if ',' not in w: logging.error("Array weights should be separated by ','") return weights = w.split(',') try: weights = list(map(int,weights)) except ValueError: logging.error("Weights has non integral value") return if ',' not in im: logging.error("Array impacts should be separated by ','") return impact = im.split(',') # outfile = args[4] for x in impact: if x != '+' and x != '-': logging.error("Impact must contain only '+' or '-'") return try: df = pd.read_csv(fname) except FileNotFoundError: logging.error("Input file provided not found") return if(len(df.columns) < 3): logging.error("Less Number of columns in Input File") return ndf = df.drop('Fund Name', axis=1) row = len(ndf) cols = len(ndf.iloc[0,:]) for i in range(row): rows = list(ndf.iloc[i]) for j in range(cols): try: rows[j] = pd.to_numeric(rows[j]) except ValueError: logging.warning(f"Non numeric value encountered in input.csv at {i}th row and {j}th coln") if(cols != len(weights) or cols != len(impact)): logging.error("Length of inputs not match.") return den = ndf.apply(np.square).apply(np.sum,axis = 0).apply(np.sqrt) for i in range(cols): for j in range(row): ndf.iat[j, i] = (ndf.iloc[j, i] / den[i]) * weights[i] ideal_best = (ndf.max().values) ideal_worst = (ndf.min().values) for i in range(cols): if impact[i] == '-': ideal_best[i], ideal_worst[i] = ideal_worst[i], ideal_best[i] score = [] for i in range(row): p,n = 0, 0 for j in range(cols): p += (ideal_best[j] - ndf.iloc[i, j])**2 n += (ideal_worst[j] - ndf.iloc[i, j])**2 p, n = p**0.5, n**0.5 score.append(n/(p + n)) df['Topsis Score'] = score df['Rank'] = (df['Topsis Score'].rank(method='max', ascending=False)) df.to_csv(outfile)