scs-rs 0.1.0

Bindings to the SCS convex optimization solver
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188

import cvxpy
import random
import numpy
from scipy import sparse

(x, y, z) = (cvxpy.Variable(name="x", shape=3), cvxpy.Variable(name="y", shape=3), cvxpy.Variable(name="z", shape=3))

ellipsoids = [ (((1, 0, 0), (2.,0.,0.), 0.0), -5.)]#((1, 0, 0), (3.,2.,5.), 0.7), ((1, 0, 0), (5.,6.,7.), 6.0), ((1, 0, 0), (5.,6.,7.), 0.0), ((1, 0, 0), (5.,6.,7.), 0.0)
#]*10#(((1, 0, 0), [2.,3.,4.], 0.0), ((1, 0, 0), [5.,6.,7.], 0.0), -50.),
#(((1, 1, 0), [2.,3.,4.], 0.0), ((1, 0, -1), [5.,6.,7.], 0.0),((1, 0, 0), [5.,6.,7.], 0.0), -50.),
#(((1, 0, 0), [8.,9.,10.], 1.5), ((0, 1, 0), [14,15.,16.], 1.7), ((1, -1, 1), [17.,18.,19.], 1.6), -70.)]


a = cvxpy.Variable(shape=2)
problem = cvxpy.Problem(cvxpy.Maximize(a[0]), [cvxpy.norm(a + [2,0]) <= 50])

probdata, chain, inverse_data = problem.get_problem_data(cvxpy.SCS)
soln = chain.solve_via_data(problem, probdata)
print(problem)
print(probdata['A'].todense())
print(probdata['A'].__dict__)
print(probdata['b'])
print(probdata['c'])
cone_dims = probdata['dims']
cones = {
    "f": cone_dims.zero,
    "l": cone_dims.nonpos,
    "q": cone_dims.soc,
    "ep": cone_dims.exp,
    "s": cone_dims.psd,
}

print(cones)
print(soln)

exit()

objective = cvxpy.Maximize(808)

constraints = []
seen_foci = {}

for e in ellipsoids:
    expr = e[-1]
    for focus in e[:-1]:
        if focus in seen_foci:
            expr += seen_foci[focus]
            continue

        seen_foci[focus] = cvxpy.Variable(name="f" + str(len(seen_foci)))
        expr += seen_foci[focus]

        nonzero = False
        mom = 0
        # do custom simplifications
        for sign, v in zip(focus[0], (x, y, z)):
            if sign != 0:
                r = v if sign == 1 else -v
                if not nonzero:
                    nonzero = True
                    mom = r
                else:
                    if sign == 1:
                        mom += v
                    else:
                        mom -= v

        if focus[2] != 0.:
            constraints.append(cvxpy.norm(cvxpy.hstack([focus[2], mom + list(focus[1])]), 2) <= seen_foci[focus])
        else:
            constraints.append(cvxpy.norm(mom + list(focus[1]), 2) <= seen_foci[focus])

    constraints.append(expr <= 0)


# manually predict A and b
b = [-e[-1] for e in ellipsoids] + [b for a in [[0.] + ([f[2]] if f[2] != 0. else []) + list(f[1]) for e in ellipsoids for f in e[:-1]] for b in a]

varcount = sum(1 if any(f[0][i] != 0 for e in ellipsoids for f in e[:-1]) else 0 for i in range(3))
width = 3 * varcount + sum(len(e[:-1]) for e in ellipsoids)
a = [ [0. for _ in range(width)] for _ in range(len(b))]

# set the first row: first entry 1 , then last entries to 1


focus_row_counter = len(ellipsoids)
for row_a, e in enumerate(ellipsoids):
    pos = 3 * varcount + sum(len(ee[:-1]) for ee in ellipsoids[:row_a])
    range_len = len(e[:-1])

    for ii in range(range_len):
        a[row_a][pos + ii] = 1.0
    
    for foc_index, (variables, shift, mass) in enumerate(e[:-1]):
        a[focus_row_counter][pos + foc_index] = -1.

        focus_row_counter += 1
        if mass != 0.:
            # mass is an empty line
            focus_row_counter += 1

        for dir_index in range(3):
            for var_index, v in enumerate(variables):
                if v != 0:
                    a[focus_row_counter][3 * var_index + dir_index] = -float(v)

            focus_row_counter += 1



problem = cvxpy.Problem(objective, constraints)

probdata, chain, inverse_data = problem.get_problem_data(cvxpy.SCS)
soln = chain.solve_via_data(problem, probdata)
print(soln)

print(probdata['A'].todense())
print(probdata['b'])
print(problem)

import time

accumtime = 0
N = 1
for _ in range(N):
    start = time.time()
    problem.solve(cvxpy.SCS, verbose=True)#, eps=1e-2)
    print(x.value, y.value, z.value)
    end = time.time()
    #print('time', end - start)
    accumtime += end - start

print('avg time', accumtime / N)

"""
#print('v', x.value)
probdata, chain, inverse_data = problem.get_problem_data(cvxpy.SCS)
#print(probdata)
#print(probdata['G'].todense())  # for ECOS and CVXOPT

import scs
data = {
  'A': probdata['A'],
  'b': probdata['b'],
  'c': probdata['c'],
}
cone_dims = probdata['dims']
cones = {
    "f": cone_dims.zero,
    "l": cone_dims.nonpos,
    "q": cone_dims.soc,
    "ep": cone_dims.exp,
    "s": cone_dims.psd,
}

print(cones)

print('width of A', data['A'].shape, (len(a), len(a[0])))

#assert((len(a), len(a[0])) == data['A'].shape)

for aa in numpy.array(a).tolist():
    print(aa)
print("")
for aa in data['A'].todense().tolist():
    print(aa)

print('a recons=\n{}'.format(numpy.array(a).tolist()))
print('a=\n{}'.format(data['A'].todense().tolist()))
print('b=', data['b'])

#assert(numpy.array_equal(a, data['A'].todense()))
#assert(numpy.array_equal(b, probdata['b']))

# if we override the matrix, we create the wrong mapping
#probdata['A'] = sparse.csc_matrix(numpy.array(a))


# calls SCS using `data`
soln = chain.solve_via_data(problem, probdata)
print(soln)
# unpacks the solution returned by SCS into `problem`
problem.unpack_results(soln, chain, inverse_data)

print('val', x.value)
"""

#print(problem)