scipy.optimize.minimize with constraints does not work as expected - scipy-optimize-minimize

I would like to find optimal solution for the problem discribed in the attached figure https://i.stack.imgur.com/W6fhf.jpg
My code is
from scipy.optimize import minimize
def create_constraints(t, tau, ub):
constraints = [{
'type': 'ineq',
'fun': lambda x: x[0] * (-1),
}]
con1 = {
'type': 'ineq',
'fun': lambda x: x[0] - x[1] + tau[0],
}
constraints.append(con1)
con2 = {
'type': 'ineq',
'fun': lambda x: x[1] - ub + tau[1],
}
constraints.append(con2)
return constraints
def f(x):
return (x[0] + x[1]) * (-1)
t0 = np.array([1, 10])
tau0 = [5, 5]
cons = create_constraints(t0, tau0, 30)
res = minimize(f, t0, constraints=cons, method='trust-constr')
with the above parameters for constraints I expect that optimal solution would be close to [20, 25]. However the result of optimization is x: array([ 6.66666669, 18.33333334]). What am I doing wrong? why minimizer does not work as expected?
Any help appreciated.

I found that if I create constraints in this way
def create_constraints_2(t, tau, ub):
constr_f = lambda x: np.array([
x[0] * (-1),
x[0] - x[1] + tau[0],
x[1] - ub + tau[1],
])
return NonlinearConstraint(constr_f, -np.inf, 0)
And then call the functions
cons_2 = create_constraints_2(t0, tau0, 30)
minimize(f, t0, constraints=cons_2, method='trust-constr')
Then it works fine. The output is x: array([19.99974476, 24.99993157])

Related

Finding the total weight in MST in Prim's Algorithm

currently I'm making a comparison between the Prim's Algorithm and Kruskal's Algorithm. Both codes are from GeeksforGeeks, however only the Kruskal's algorithm has the total calculated weight in finding the MST. The Prim's algorithm doesn't have one, and I don't have any idea on how can I output the total weight. I hope you can help me.
Here's the code for the Kruskal's Algorithm (from GeeksforGeeks):
class Graph:
def __init__(self, vertices):
self.V = vertices
self.graph = []
def addEdge(self, u, v, w):
self.graph.append([u, v, w])
def find(self, parent, i):
if parent[i] == i:
return i
return self.find(parent, parent[i])
def union(self, parent, rank, x, y):
xroot = self.find(parent, x)
yroot = self.find(parent, y)
if rank[xroot] < rank[yroot]:
parent[xroot] = yroot
elif rank[xroot] > rank[yroot]:
parent[yroot] = xroot
else:
parent[yroot] = xroot
rank[xroot] += 1
def KruskalMST(self):
result = []
i = 0
e = 0
self.graph = sorted(self.graph,
key=lambda item: item[2])
parent = []
rank = []
for node in range(self.V):
parent.append(node)
rank.append(0)
while e < self.V - 1:
u, v, w = self.graph[i]
i = i + 1
x = self.find(parent, u)
y = self.find(parent, v)
if x != y:
e = e + 1
result.append([u, v, w])
self.union(parent, rank, x, y)
minimumCost = 0
print("Edges in the constructed MST")
for u, v, weight in result:
minimumCost += weight
print("%d -- %d == %d" % (u, v, weight))
print("Minimum Spanning Tree", minimumCost)
g = Graph(4)
g.addEdge(0, 1, 10)
g.addEdge(0, 2, 6)
g.addEdge(0, 3, 5)
g.addEdge(1, 3, 15)
g.addEdge(2, 3, 4)
g.KruskalMST()
The code for Prim's Algorithm (also from GeeksforGeeks):
import sys
class Graph():
def __init__(self, vertices):
self.V = vertices
self.graph = [[0 for column in range(vertices)]
for row in range(vertices)]
minimumcost = 0
def printMST(self, parent):
print ("Edge \tWeight")
for i in range(1, self.V):
print (parent[i], "-", i, "\t", self.graph[i][parent[i]])
def minKey(self, key, mstSet):
min = sys.maxsize
for v in range(self.V):
if key[v] < min and mstSet[v] == False:
min = key[v]
min_index = v
return min_index
def primMST(self):
key = [sys.maxsize] * self.V
parent = [None] * self.V
key[0] = 0
mstSet = [False] * self.V
parent[0] = -1
for cout in range(self.V):
u = self.minKey(key, mstSet)
mstSet[u] = True
for v in range(self.V):
if self.graph[u][v] > 0 and mstSet[v] == False and key[v] > self.graph[u][v]:
key[v] = self.graph[u][v]
parent[v] = u
self.printMST(parent)
g = Graph(5)
g.graph = [ [0, 2, 0, 6, 0],
[2, 0, 3, 8, 5],
[0, 3, 0, 0, 7],
[6, 8, 0, 0, 9],
[0, 5, 7, 9, 0]]
g.primMST();

I am trying to give the heuristics using random function. Can anyone please correct me

def getNeighbors(v, adjacency_list):
return adjacency_list[v]
import random
def h(n):
H = {
'A': 14,
'B': 12,
'C': 11,
'D': 6,
'E': 4,
'F': 11,
'Z':random.randint(0,22),
Error message shows in this line saying invalid syntax.
h('Z')=='Z',
}
return H[n]
def A_star(start_node, stop_node,adjacency_list):
open_list = set([start_node])
closed_list = set([])
g = {}
g[start_node] = 0
parents = {}
parents[start_node] = start_node
while len(open_list) > 0:
n = None
for v in open_list:
if n == None or g[v] + h(v) < g[n] + h(n):
n = v;
if n == None:
print('Path does not exist!')
return None
if n == stop_node:
reconst_path = []
while parents[n] != n:
reconst_path.append(n)
n = parents[n]
reconst_path.append(start_node)
reconst_path.reverse()
print('Path found: {}'.format(reconst_path))
return reconst_path
for (m, weight) in getNeighbors(n,adjacency_list):
if m not in open_list and m not in closed_list:
open_list.add(m)
parents[m] = n
g[m] = g[n] + weight
else:
if g[m] > g[n] + weight:
g[m] = g[n] + weight
parents[m] = n
if m in closed_list:
closed_list.remove(m)
open_list.add(m)
open_list.remove(n)
closed_list.add(n)
print('Path does not exist!')
return None
adjacency_list = {
'A': [('B', 4), ('C', 3)],
'B': [('E', 12),('F',5)],
'C': [('D', 7),('E',10)],
'D': [('E',2)],
'E': [('Z',5)],
'F': [('Z',16)]
}
A_star('A', 'Z',adjacency_list)
I would like to know if there is anything wrong with my code. Here I am trying to implement the A star algorithm while giving heuristics randomly, Are there any mistakes that I have done. I know there might be a very silly one. If you find anything please correct me.

Trivial change but different results (ForAll and Exists)

I have a set of inequalities that I want to find a (trivial) solution.
When I use the Exists operator, everything works great, as you can see in this Z3 script and in its Z3Py version.
#!/bin/python
from z3 import *
# we have that
s = Solver()
## mu0_px is the initial marking for place px;
mu_p1, mu_p2, mu_p3 = 0, 0, 1
## pi_tj is the pre-condition from place pi to transition tj
p1_t1, p1_t2, p1_t3 = 1, 0, 0
p2_t1, p2_t2, p2_t3 = 0, 1, 0
p3_t1, p3_t2, p3_t3 = 0, 0, 1
## tj_pi is the post-condition from transition tj to place pi
t1_p1, t2_p1, t3_p1 = 0, 1, 0
t1_p2, t2_p2, t3_p2 = 1, 0, 0
t1_p3, t2_p3, t3_p3 = 0, 0, 0
## find the values for the faulty transitions
f_p1, p1_f = Ints('f_p1 p1_f')
f_p2, p2_f = Ints('f_p2 p2_f')
f_p3, p3_f = Ints('f_p3 p3_f')
# where they should be
s.add( f_p1 == 1, f_p2 == 0, f_p3 == 0 )
s.add( p1_f == 0, p2_f == 0, p3_f == 1 )
## l \in Naturals ;
l11 = Int('l11')
# Sequence 11: t1,t2,t3
s11_t1, s11_t2, s11_t3 = 1, 1, 0
# It does works! :o
s.add( l11 == 1 )
s.add(
Exists([l11],
Or(
mu_p1 + (t1_p1-p1_t1)*s11_t1 + (t2_p1-p1_t2)*s11_t2 + (t3_p1-p1_t3)*s11_t3 + l11 * (f_p1 - p1_f) < p1_t3,
mu_p2 + (t1_p2-p2_t1)*s11_t1 + (t2_p2-p2_t2)*s11_t2 + (t3_p2-p2_t3)*s11_t3 + l11 * (f_p2 - p2_f) < p2_t3,
mu_p3 + (t1_p3-p3_t1)*s11_t1 + (t2_p3-p3_t2)*s11_t2 + (t3_p3-p3_t3)*s11_t3 + l11 * (f_p3 - p3_f) < p3_t3,
)
)
)
print(s)
print(s.check())
print(s.model())
However, when I replace the existential quantifier by Forall as in this link, and in the Python code below, there is no solution when I believe that it should still be sat.
#!/bin/python
from z3 import *
# we have that
s = Solver()
## mu0_px is the initial marking for place px;
mu_p1, mu_p2, mu_p3 = 0, 0, 1
## pi_tj is the pre-condition from place pi to transition tj
p1_t1, p1_t2, p1_t3 = 1, 0, 0
p2_t1, p2_t2, p2_t3 = 0, 1, 0
p3_t1, p3_t2, p3_t3 = 0, 0, 1
## tj_pi is the post-condition from transition tj to place pi
t1_p1, t2_p1, t3_p1 = 0, 1, 0
t1_p2, t2_p2, t3_p2 = 1, 0, 0
t1_p3, t2_p3, t3_p3 = 0, 0, 0
## find the values for the faulty transitions
f_p1, p1_f = Ints('f_p1 p1_f')
f_p2, p2_f = Ints('f_p2 p2_f')
f_p3, p3_f = Ints('f_p3 p3_f')
# where they should be
s.add( f_p1 == 1, f_p2 == 0, f_p3 == 0 )
s.add( p1_f == 0, p2_f == 0, p3_f == 1 )
## l \in Naturals ;
l11 = Int('l11')
# Sequence 11: t1,t2,t3
s11_t1, s11_t2, s11_t3 = 1, 1, 0
# It does not work! :(
s.add( l11 == 1 )
s.add(
ForAll([l11],
Or(
mu_p1 + (t1_p1-p1_t1)*s11_t1 + (t2_p1-p1_t2)*s11_t2 + (t3_p1-p1_t3)*s11_t3 + l11 * (f_p1 - p1_f) < p1_t3,
mu_p2 + (t1_p2-p2_t1)*s11_t1 + (t2_p2-p2_t2)*s11_t2 + (t3_p2-p2_t3)*s11_t3 + l11 * (f_p2 - p2_f) < p2_t3,
mu_p3 + (t1_p3-p3_t1)*s11_t1 + (t2_p3-p3_t2)*s11_t2 + (t3_p3-p3_t3)*s11_t3 + l11 * (f_p3 - p3_f) < p3_t3,
)
)
)
print(s)
print(s.check())
print(s.model())
Did anyone ever have a problem like this before?
The variable l11 as you declared and the one that gets used in the quantification are totally different: In particular, you stating it equals 1 have no bearing in the quantified formula. So you get sat with existential but unsat with universal since the formula is clearly not true for all values of l11.
This might be confusing, but it is the intended behaviour. To see the effect, simply print the smtlib equivalent and you’ll see how the variables are assigned.

Scipy.optimize - minimize not respecting constraints

Using the code below to to understand how Scipy optmization/minimization works. The results are not matching what I am expecting.
"""
Minimize: f = 2*x[0]*x[1] + 2*x[0] - x[0]**2 - 2*x[1]**2
Subject to: -2*x[0] + 2*x[1] <= -2
2*x[0] - 4*x[1] <= 0
x[0]**3 -x[1] == 0
where: 0 <= x[0] <= inf
1 <= x[1] <= inf
"""
import numpy as np
from scipy.optimize import minimize
def objective(x):
return 2.0*x[0]*x[1] + 2.0*x[0] - x[0]**2 - 2.0*x[1]**2
def constraint1(x):
return +2.0*x[0] - 2.0*x[1] - 2.0
def constraint2(x):
return -2.0*x[0] + 4.0*x[1]
def constraint3(x):
sum_eq = x[0]**3.0 -x[1]
return sum_eq
# initial guesses
n = 2
x0 = np.zeros(n)
x0[0] = 10.0
x0[1] = 100.0
# show initial objective
print('Initial SSE Objective: ' + str(objective(x0)))
# optimize
#b = (1.0,None)
bnds = ((0.0,1000.0), (1.0,1000.0))
con1 = {'type': 'ineq', 'fun': constraint1}
con2 = {'type': 'ineq', 'fun': constraint2}
con3 = {'type': 'eq', 'fun': constraint3}
cons = ([con1, con2, con3])
solution = minimize(objective,
x0,
method='SLSQP',
bounds=bnds,
constraints=cons)
x = solution.x
print(solution)
# show final objective
print('Final SSE Objective: ' + str(objective(x)))
# print solution
print('Solution')
print('x1 = ' + str(x[0]))
print('x2 = ' + str(x[1]))
print('\n')
print('x', x)
print('constraint1', constraint1(x))
print('constraint2', constraint2(x))
print('constraint3', constraint3(x))
When I run, this is what Python throws on its output console:
Initial SSE Objective: -18080.0
fun: 2.0
jac: array([ 0.00000000e+00, -2.98023224e-08])
message: 'Optimization terminated successfully.'
nfev: 122
nit: 17
njev: 13
status: 0
success: True
x: array([2., 1.])
Final SSE Objective: 2.0
Solution
x1 = 2.0000000000010196
x2 = 1.0000000000012386
x [2. 1.]
constraint1 -4.3787196091216174e-13
constraint2 2.915001573455811e-12
constraint3 7.000000000010997
Despite the optimizer says the result was successful, the constraint3 is not respected because the result should be zero. What am I missing?
Your problem is incompatible. You can eliminate the 3rd constraint (which makes your problem simpler in the first place - only a scalar optimization), after this it is a bit more clear to see what is the problem. From constraint 3 and the lower bound on the original x1 follows, that x0 is not feasible from 0 to 1, so the lower bound in the 1D problem should be 1. It is easy to see that constraint 2 will be always positive, when x0 is larger than 1, therefore it will never be satisfied.
When I run your original problem for me it stops with positive directional derivative (and for the rewritten problem with 'Inequality constraints incompatible').
Which SciPy are you using? For me it is 1.4.1.
On the picture below you can see the objective and the remaining constraints for the 1D problem (horizontal axis is the original x0 variable)
"""
Minimize: f = 2*x[0]*x1 + 2*x[0] - x[0]**2 - 2*x1**2
Subject to: -2*x[0] + 2*x[1] <= -2
2*x[0] - 4*x[1] <= 0
x[0]**3 -x[1] == 0
where: 0 <= x[0] <= inf
1 <= x[1] <= inf
"""
import numpy as np
from scipy.optimize import minimize
def objective(x):
return 2*x**4 + 2*x - x**2 - 2*x**6
def constraint1(x):
return x - x**3 - 1
def constraint2(x):
return 2 * x**3 - x
#
# def constraint3(x):
# sum_eq = x[0]**3.0 -x[1]
# return sum_eq
# initial guesses
n = 1
x0 = np.zeros(n)
x0[0] = 2.
# x0[1] = 100.0
# show initial objective
print('Initial SSE Objective: ' + str(objective(x0)))
# optimize
#b = (1.0,None)
bnds = ((1.0,1000.0),)
con1 = {'type': 'ineq', 'fun': constraint1}
con2 = {'type': 'ineq', 'fun': constraint2}
# con3 = {'type': 'eq', 'fun': constraint3}
cons = [
# con1,
con2,
# con3,
]
solution = minimize(objective,
x0,
method='SLSQP',
bounds=bnds,
constraints=cons)
x = solution.x
print(solution)
# show final objective
print('Final SSE Objective: ' + str(objective(x)))
# print solution
print('Solution')
print('x1 = ' + str(x[0]))
# print('x2 = ' + str(x[1]))
print('\n')
print('x', x)
print('constraint1', constraint1(x))
print('constraint2', constraint2(x))
# print('constraint3', constraint3(x))
x_a = np.linspace(1, 2, 200)
f = objective(x_a)
c1 = constraint1(x_a)
c2 = constraint2(x_a)
import matplotlib.pyplot as plt
plt.figure()
plt.plot(x_a, f, label="f")
plt.plot(x_a, c1, label="c1")
plt.plot(x_a, c2, label="c2")
plt.legend()
plt.show()

ArithGeo(arr) CoderByte Ruby: Why doesn't this solution work for certain test case scenarios

Below I have posted the instructions for this problem along with my solution. A few test case scenarios have failed, but seem to be working for most. Can anybody help out at which point I've gone wrong? Any help is much appreciated!!
Using the Ruby language, have the function ArithGeo(arr) take the array of numbers stored in arr and return the string "Arithmetic" if the sequence follows an arithmetic pattern or return "Geometric" if it follows a geometric pattern.
If the sequence doesn't follow either pattern return -1.
An arithmetic sequence is one where the difference between each of the numbers is consistent
Arithmetic example: [2, 4, 6, 8]
In a geometric sequence, each term after the first is multiplied by some constant or common ratio.
Geometric example: [2, 6, 18, 54]
Negative numbers may be entered as parameters, 0 will not be entered, and no array will contain all the same elements.
Code:
def arithGeo(num)
idx = 0
while idx < num.length
if ((num[idx] - num[idx + 1]) == (num[idx + 1] - num[idx + 2]))
return "Arithmetic"
elsif ((num[idx + 1] / num[idx]) == (num[idx + 2] / num[idx + 1]))
return "Geometric"
else
return "-1"
end
idx += 1
end
end
#Test Cases that Failed
p arithGeo([1, 2, 3, 4, 5, 10, 20])
p arithGeo([1, 2, 3, 4, 5, 6, 7, 88, 2])
p arithGeo([10, 110, 210, 310, 410, 511])
OK, lets do a much more "ruby like" way:
def arith?(arr)
check_arr = []
arr.reverse.inject {|memo, num| check_arr << (memo - num); num}
#loop through from highest to lowest, subtracting each from the next and store in check_arr
check_arr.all? {|num| num == check_arr[-1]}
#check that all results are the same in the arr i.e. [2,2,2,2,2]
end
This returns true if all of the operations return the same result, thus a linear progression.
def geo?(arr)
check_arr = []
arr.reverse.inject {|memo, num| check_arr << (memo / num); num}
#loop through from highest to lowest, dividing each by the next and store in check_arr
check_arr.all? {|x| x == check_arr[-1]}
#check that all products are the same in the arr i.e. [3,3,3,3,3]
end
This returns true if all of the operations return the same result, thus a geometric progression.
Now use those methods in your other method
def arith_geo?(arr)
if arith?(arr)
'Arithmetic'
elsif geo?(arr)
'Geometric'
else
-1
end
end
You did use a while but you do not loop over the data, because you write return you will only ever look at the first three numbers and then immediately return the result. You will have to keep the previous result, and make sure the result stays the same to return either geometric or arithmetic.
This should help you to complete the exercise :)
I was able to do the solution in JavaScript and this is what I came up with:
function algoGeo(arr){
var algo = true;
var geo = true;
//first check algo
for(var k = 1; k < arr.length; k++){
if( (arr[0] + (arr[1] - arr[0]) * k) !== arr[k] && algo ){
algo = false;
}
if( arr[0] * Math.pow(arr[1] / arr[0], k) !== arr[k] && geo){
geo = false;
}
}
return algo ? "Arithmetic" : geo ? "Geometric" : -1;
}
var arr = [5,12,19,26];
console.log(algoGeo(arr));
def ArithGeo(arr)
diff1 = []
diff2 = []
arr.each_index do |x|
if(x + 1 < arr.length)
diff1 << arr[x + 1] - arr[x]
diff2 << arr[x + 1] / arr[x]
end
end
diff1.uniq.size == 1 ? "Arithmetic" : diff2.uniq.size == 1 ? "Geometric" : -1
end
A little late but this is what i came up with when trying to solve this same question.

Resources