binning.py

"""
How to Handle Noisy Data in preprocessing of data?
	Binning method
"""
import math
import random
import statistics

size = int(input("enter the size : "))
numbers = [random.randrange(100) for i in range(size)]
# sorting
numbers = sorted(numbers)
print("chosen numbers : ",numbers)

bins = int(input("How many bins(input should divide the size perfectly)? : "))
if size%bins == 0:
	# number of item in each bin
	c = int(size/bins)
	# Partition into (equi-depth) bins
	equi_depth = [[numbers[p] for p in range(i,i+c)] for i in range(0,size,c) ]
	# Smoothing by bin means
	smooth_bin_means = [[statistics.mean(i) for j in range(c)] for i in equi_depth]
	# Smoothing by bin boundaries
	smooth_bin_boundary = []
	for i in equi_depth:
		min_num = i[0]
		max_num = i[-1]
		item = []
		for j in i:
			if j - min_num <= max_num - j:
				item.append(min_num)
			else:
				item.append(max_num)
		smooth_bin_boundary.append(item)

	print("Partition into (equi-depth) bins : ",equi_depth)
	print("Smoothing by bin means : ",smooth_bin_means)
	print("Smoothing by bin boundaries : ",smooth_bin_boundary)
else:
	print("incorrect input")