/
testA11.py
126 lines (107 loc) · 4.18 KB
/
testA11.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
import imageIO
import numpy
import a11
import tests
from halide import *
def main():
im=imageIO.imread('rgb-small.png')
lumi=im[:,:,1] #I'm lazy, I'll just use green
smallLumi=numpy.transpose(lumi[0:6, 0:6])
# Replace if False: by if True: once you have implement the required functions.
# Exercises:
if True:
outputNP, myFunc=a11.smoothGradientNormalized()
print ' Dimensionality of Halide Func:', myFunc.dimensions()
imageIO.imwrite(outputNP, 'normalizedGradient.png')
if True:
outputNP, myFunc=a11.wavyRGB()
print ' Dimensionality of Halide Func:', myFunc.dimensions()
imageIO.imwrite(outputNP, 'rgbWave.png')
if True:
outputNP, myFunc = a11.luminance(im)
print ' Dimensionality of Halide Func:', myFunc.dimensions()
imageIO.imwrite(outputNP, 'rgbLuminance.png')
imageIO.imwrite(tests.luminance(im), 'rgbLuminancePython.png')
if True:
outputNP, myFunc=a11.sobel(lumi)
print ' Dimensionality of Halide Func:', myFunc.dimensions()
imageIO.imwrite(outputNP, 'sobelMag.png')
imageIO.imwrite(tests.sobelMagnitude(lumi), 'sobelMagPython.png')
if False:
L=a11.pythonCodeForBoxSchedule5(smallLumi)
for x in L:
print x
print ""
if False:
L=a11.pythonCodeForBoxSchedule6(smallLumi)
print "Schedule 6:"
for x in L:
print x
print ""
if False:
L=a11.pythonCodeForBoxSchedule7(smallLumi)
print "Schedule 7"
for x in L:
print x
print ""
if True:
outputNP, myFunc=a11.localMax(lumi)
print ' Dimensionality of Halide Func:', myFunc.dimensions()
imageIO.imwrite(outputNP, 'maxi.png')
if False:
testWhite = numpy.ones((50,50))
input = Image(Float(32), testWhite)
x, y = Var('x'), Var('y')
clamped = Func('clamped')
clamped[x, y] = input[clamp(x, 0, input.width()-1),
clamp(y, 0, input.height()-1)]
sigma = 2.0 # IS THIS RIGHT?!?
blurX, finalBlur= a11.GaussianSingleChannel(clamped , sigma, trunc=3)
blurXOutput = blurX.realize(input.width(), input.height())
blurXNP = numpy.array(Image(blurXOutput))
imageIO.imwrite(blurXNP, 'blurXWhite.png')
if True:
input=Image(Float(32), lumi)
x, y = Var('x'), Var('y')
clamped = Func('clamped')
clamped[x, y] = input[clamp(x, 0, input.width()-1),
clamp(y, 0, input.height()-1)]
sigma = 5.0 # IS THIS RIGHT?!?
blurX, finalBlur= a11.GaussianSingleChannel(clamped , sigma, trunc=3)
blurXOutput = blurX.realize(input.width(), input.height())
blurXNP = numpy.array(Image(blurXOutput))
imageIO.imwrite(blurXNP, 'blurX.png')
finalBlurOutput = finalBlur.realize(input.width(), input.height())
finalBlurNP = numpy.array(Image(finalBlurOutput))
imageIO.imwrite(finalBlurNP, 'finalBlur.png')
if True:
# im=numpy.load('Input/hk.npy')
# outputNP, myFunc=a11.harris(im, 0)
# imageIO.imwrite(outputNP, 'harris.png')
outputNP, myFunc=a11.harris(im, 1)
imageIO.imwrite(outputNP, 'harris_fast.png')
print ' Dimensionality of Halide Func:', myFunc.dimensions()
if False:
# Timing for Harris
im=imageIO.imread('hk.png')
myFunc = a11.harris(im, 0)
runAndMeasure(myFunc, im.shape[1], im.shape[0])
myFunc = a11.harris(im, 1)
runAndMeasure(myFunc, im.shape[1], im.shape[0])
imageIO.imwrite(resultNP, 'harrisFast.png')
def runAndMeasure(myFunc, w, h, nTimes=5):
L=[]
output=None
myFunc.compile_jit()
for i in xrange(nTimes):
t=time.time()
output = myFunc.realize(w,h)
L.append (time.time()-t)
hIm=Image(output)
mpix=hIm.width()*hIm.height()/1e6
print 'best: ', numpy.min(L), 'average: ', numpy.mean(L)
print '%.5f ms per megapixel (%.7f ms for %.2f megapixels)' % (numpy.mean(L)/mpix*1e3, numpy.mean(L)*1e3, mpix)
return numpy.array(hIm)
#usual python business to declare main function in module.
if __name__ == '__main__':
main()