CODEWITHSUNDEEP
pythonnumpyinventorymdpmdptoolbox
Chris
Chris

Reputation: 95

Python returning two identical matrices

I am trying to write a small program for Markov Decision Process (inventory problem) using Python. I cannot figure out why the program outputs two identical matrices (for profit and decision matrices). The programming itself has some problems too because the last two columns are all zeros, which should not happen. Any help on the program itself would also be super helpful!

import math
import numpy as np

def salvageCost(b):
    return 5 * b

def orderingCost(b):
    if b == 0:
        return 0
    else:
        return 4 + 2 * b

def holdingCost(b):
    return 1.5 * b

def revenue(b):
    return 8 * b

M = 10
N = 4
u = np.zeros((M+1,N))
T = np.array([4,3,2,1])
S = np.array(range(M+1))
A = np.array(range(M+1))
u[:,0] = S*5
d = np.zeros((11,4))

probs = np.array([0.05, 0.1, 0.15, 0.2, 0.2, 0.15, 0.1, 0.05, 0, 0, 0])
demands = np.array(range(11))
candidate = [0]*11
d = u

for i in T[1:N]:
    for j in S:
        for a in range(M-j+1):
            candidate[a] = -holdingCost(j) - orderingCost(a) \
            + np.array([probs[k]for k in range(j+a)]).dot(8*np.array([demands[k]for k in range(j+a)])) \
            + np.array([probs[p] for p in range(min(j+a+1,M+1),M+1)]).dot(np.array(range(min(j+a+1,M+1),M+1)))*80\
            + probs.dot(u[:,i-1])
        u[j,i] = max(candidate)
        d[j,i] = candidate.index(max(candidate))
    


print(d)
print(u)

This is the output

[[ 0.  0.  0.  0.]
 [ 5.  0.  0.  0.]
 [10.  0.  0.  0.]
 [15.  0.  0.  0.]
 [20.  0.  0.  0.]
 [25.  0.  0.  0.]
 [30.  0.  0.  0.]
 [35.  0.  0.  0.]
 [40.  0.  0.  0.]
 [45.  0.  0.  0.]
 [50.  0.  0.  0.]]
[[ 0.  0.  0.  0.]
 [ 5.  0.  0.  0.]
 [10.  0.  0.  0.]
 [15.  0.  0.  0.]
 [20.  0.  0.  0.]
 [25.  0.  0.  0.]
 [30.  0.  0.  0.]
 [35.  0.  0.  0.]
 [40.  0.  0.  0.]
 [45.  0.  0.  0.]
 [50.  0.  0.  0.]]

Upvotes: 1

Views: 66

Answers (2)

BrokenBenchmark
BrokenBenchmark

Reputation: 19253

This line:

d = u

points d and u to the same location in memory. As a result, any changes to d are reflected in u, and vice versa.

You are probably looking to create a copy of the contents of u, in which case you can use d = u.copy().

Upvotes: 2

Kraigolas
Kraigolas

Reputation: 5590

Because numpy.ndarrays are mutable, when you write

d = u

d and u point to the same matrix, so any change to d or to u affects both:

>>> import numpy as np
>>> x = np.array([1, 2])
>>> y = x
>>> y[0] = 10
>>> x
array([10,  2])

Upvotes: 2

Related Questions