-
Notifications
You must be signed in to change notification settings - Fork 0
/
test_GMM.py
170 lines (134 loc) · 3.96 KB
/
test_GMM.py
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
import Image
import time
from GMM import GMM
from sklearn import mixture
from clutil import padArray2D
from clutil import roundUp
import numpy as np
import pyopencl as cl
cm = cl.mem_flags
def calcA_cpu(weights, means, covars, A=None):
nComp = weights.shape[0]
if A == None:
ret = True
A = np.empty((nComp, 8), np.float32)
else:
ret = False
for i in xrange(0, nComp):
w = np.log(weights[i])
h = -0.5*( 3*np.log(2*np.pi) + np.sum(np.log(covars[i])) + np.sum((means[i]**2) / (covars[i])) )
A[i][0] = w + h
A[i][1:4] = (means[i]/covars[i])
A[i][4] = 0
A[i][5:8] = -1/(2*covars[i])
if ret:
return A
platforms = cl.get_platforms()
devices = platforms[0].get_devices()
device = devices[1]
context = cl.Context([device])
queue = cl.CommandQueue(context,properties=cl.command_queue_properties.PROFILING_ENABLE)
np.set_printoptions(suppress=True)
src = Image.open("/Users/marcdeklerk/msc/code/dataset/processed/source/800x600/GT04.png")
if src.mode != 'RGBA':
src = src.convert('RGBA')
width = src.size[0]
height = src.size[1]
shape = src.size
shapeNp = (src.size[1], src.size[0])
size = width * height
lWorksize = (16, 16)
hSrc = padArray2D(np.array(src).view(np.uint32).squeeze(), roundUp(shapeNp, lWorksize), 'edge')
dSrc = cl.Buffer(context, cm.READ_ONLY | cm.COPY_HOST_PTR, hostbuf=hSrc)
rgb = hSrc.reshape(-1, 1).view(np.uint8).astype(np.float32)[:, 0:3]
x0 = 0
x1 = 500
y0 = 200
y1 = 220
rect = (y1-y0, x1-x0)
nSamples = rect[0]*rect[1]
nIter = 65
nComp = 4
nDim = 3
samples = hSrc[y0:y1, x0:x1].ravel()
dSamples = cl.Buffer(context, cm.READ_ONLY | cm.COPY_HOST_PTR, hostbuf=samples)
samples = samples.reshape(nSamples, 1).view(np.uint8).astype(np.float32)[:, 0:3]
weights = np.tile(1.0 / nComp, nComp).astype(np.float32)
i = np.random.randint(0, nSamples-1, nComp)
means = samples[i]
covars = np.empty((nComp, 3), np.float32)
covars[:] = 10
gmm_cpu = mixture.GMM(nComp)
gmm_cpu.dtype = np.float32
gmm_cpu.init_params = ''
gmm_cpu.means_ = means
gmm_cpu.weights_ = weights
gmm_cpu.covars_ = covars
gmm_cpu.fit(samples)
gmm = GMM(context, nIter, nComp, nSamples)
a = calcA_cpu(weights, means, covars)
cl.enqueue_copy(queue, gmm.dA, a).wait()
gmm.has_preset_wmc = True
w,m,c = gmm.fit(dSamples, nSamples, retParams=True)
print 'converged: {0}'.format(gmm.has_converged)
print gmm_cpu.weights_
print w
print
print gmm_cpu.means_
print m
print
print gmm_cpu.covars_
print c
gmm_cpu.init_params = 'wmc'
iter = 10
#to estimate wmc on cpu
elapsed = 0
t1 = t2 = 0
for i in xrange(iter):
t1 = time.time()
gmm_cpu.fit(samples)
#a = calcA_cpu(gmm_cpu.weights_, gmm_cpu.means_, gmm_cpu.covars_)
#cl.enqueue_copy(clQueue, gmm.dA, a).wait()
#gmm.score(dSrc, dOut)
elapsed += time.time()-t1
print elapsed/iter
hOut = np.empty((hSrc.shape), np.float32)
dOut = cl.Buffer(context, cm.READ_WRITE | cm.COPY_HOST_PTR, hostbuf=hOut)
elapsed = 0
t1 = t2 = 0
for i in xrange(iter):
gmm.has_converged = False
t1 = time.time()
#gmm.fit(samples)
w,m,c = gmm.fit(dSamples, nSamples, retParams=True)
#gmm.fit(dOut, nSamples)
#gmm.score(dSrc, dOut)
elapsed += time.time()-t1
print elapsed/iter
#to estimate wmc for data already on gpu
elapsed = 0
t1 = t2 = 0
for i in xrange(iter):
t1 = time.time()
gmm_cpu.fit(samples)
#a = calcA_cpu(gmm_cpu.weights_, gmm_cpu.means_, gmm_cpu.covars_)
#cl.enqueue_copy(clQueue, gmm.dA, a).wait()
#gmm.score(dSrc, dOut)
elapsed += time.time()-t1
print 'cpu fit: ', elapsed/iter
hOut = np.empty((hSrc.shape), np.float32)
dOut = cl.Buffer(context, cm.READ_WRITE | cm.COPY_HOST_PTR, hostbuf=hOut)
elapsed = 0
t1 = t2 = 0
for i in xrange(iter):
t1 = time.time()
gmm.has_converged = False
#gmm.fit(samples)
# w,m,c = gmm.fit(queue, samples, nSamples, weights, means, covars, retParams=True)
gmm.fit(dSamples, nSamples)
#gmm.fit(dOut, nSamples)
#gmm.score(dSrc, dOut)
elapsed += time.time()-t1
print 'gpu fit: ', elapsed/iter
dScore = cl.Buffer(context, cm.READ_WRITE, 4*nSamples)
True