def classify_image(self, img, normMean=None, norm_std=None):
start_time = time.clock()
row_range = 1
img = normalizeImage(img)
imSize = np.shape(img)
membraneProbabilities = np.zeros(np.shape(img))
patchSize = np.int(np.sqrt(self.hiddenLayers[0].W.eval().shape[0]))
data_shared = shared_single_dataset(np.zeros((imSize[0]*row_range,patchSize**2)), borrow=True)
classify = theano.function(
[],
self.p_y_given_x,
givens={x: data_shared}
)
for row in xrange(0,1024,row_range):
if row%100 == 0:
print row
data = generate_patch_data_rows(img, rowOffset=row, rowRange=row_range, patchSize=patchSize, imSize=imSize, data_mean=normMean, data_std=norm_std)
data_shared.set_value(np.float32(data))
result = classify()
membraneProbabilities[row,:] = result[:,1]
end_time = time.clock()
total_time = (end_time - start_time)
print >> sys.stderr, ('Running time: ' +
'%.2fm' % (total_time / 60.))
return np.array(membraneProbabilities)
def predict_image(self, x, img, normMean=None, norm_std=None):
start_time = time.clock()
row_range = 1
img = normalizeImage(img)
imSize = np.shape(img)
membraneProbabilities = np.zeros(1024*1024, dtype=int )
patchSize = np.int(np.sqrt(self.hiddenLayers[0].W.eval().shape[0]))
data_shared = shared_single_dataset(np.zeros((imSize[0]*row_range,patchSize**2)), borrow=True)
classify = theano.function(
[],
self.logRegressionLayer.y_pred,
givens={x: data_shared}
)
for row in xrange(0,1024,row_range):
if row%100 == 0:
print row
data = generate_patch_data_rows(img, rowOffset=row, rowRange=row_range, patchSize=patchSize, imSize=imSize, data_mean=normMean, data_std=norm_std)
data_shared.set_value(np.float32(data))
membraneProbabilities[row*1024:row*1024+row_range*1024] = classify()
end_time = time.clock()
total_time = (end_time - start_time)
print >> sys.stderr, ('Running time: ' +
'%.2fm' % (total_time / 60.))
return np.array(membraneProbabilities)
def debug_whole_image_classification(self, img, normMean=None, norm_std=None):
start_time = time.clock()
row_range = 1
img = normalizeImage(img)
imSize = np.shape(img)
membraneProbabilities = np.zeros(np.shape(img))
patchSize = self.patchSize
data_shared = shared_single_dataset(np.zeros((imSize[0]*row_range,patchSize**2)), borrow=True)
classify = theano.function(
[],
self.debug_x,
givens={x: data_shared}
)
for row in xrange(0,33,row_range):
if row%100 == 0:
print row
data = generate_patch_data_rows(img, rowOffset=row, rowRange=row_range, patchSize=patchSize, imSize=imSize, data_mean=normMean, data_std=norm_std)
data_shared.set_value(np.float32(data))
result = classify()
#membraneProbabilities[row,:] = result[:,1]
membraneProbabilities = result
end_time = time.clock()
total_time = (end_time - start_time)
print "Image classification took %f seconds" % (total_time)
return np.array(membraneProbabilities)
def classify(self, image, mean=None, std=None):
#imSize = (1024,1024)
imSize = np.shape(image)
print 'imSize:', imSize
image = Utility.pad_image( image, self.patchSize )
imSizePadded = np.shape(image)
print 'mlp.classify mean:', mean, 'std:', std
start_time = time.clock()
row_range = 1
#imSize = np.shape(image)
#membraneProbabilities = np.zeros(np.shape(image))
membraneProbabilities = np.zeros(imSize)
patchSize = np.int(np.sqrt(self.hiddenLayers[0].W.eval().shape[0]))
print 'imSize:', imSize
print 'imSizePadded:', imSizePadded
print 'mp:', np.shape(membraneProbabilities)
data_shared = shared_single_dataset(np.zeros((imSize[0]*row_range,patchSize**2)), borrow=True)
classify = theano.function(
[],
self.p_y_given_x,
givens={self.x: data_shared}
)
for row in xrange(0,1024,row_range):
if row%100 == 0:
print row
#data = Utility.get_patch(image, row, row_range, patchSize )
#print 'data shape:', data.shape
'''
data = Utility.generate_patch_data_rows(
image,
rowOffset=row,
rowRange=row_range,
patchSize=patchSize,
imSize=imSizePadded,
data_mean=mean,
data_std=std)
'''
data = Utility.get_patch(
image,
row,
row_range,
patchSize,
data_mean=mean,
data_std=std)
#print 'data:', np.shape(data)
data_shared.set_value(np.float32(data))
result = classify()
#print 'results:', np.shape(result)
membraneProbabilities[row,:] = result[:,0]
end_time = time.clock()
total_time = (end_time - start_time)
print >> sys.stderr, ('Running time: ' +
'%.2fm' % (total_time / 60.))
return np.array(membraneProbabilities)
def classify(self, image, mean=None, std=None):
#imSize = (1024,1024)
imSize = np.shape(image)
print 'imSize:', imSize
image = Utility.pad_image(image, self.patchSize)
imSizePadded = np.shape(image)
print 'mlp.classify mean:', mean, 'std:', std
start_time = time.clock()
row_range = 1
#imSize = np.shape(image)
#membraneProbabilities = np.zeros(np.shape(image))
membraneProbabilities = np.zeros(imSize)
patchSize = np.int(np.sqrt(self.hiddenLayers[0].W.eval().shape[0]))
print 'imSize:', imSize
print 'imSizePadded:', imSizePadded
print 'mp:', np.shape(membraneProbabilities)
data_shared = shared_single_dataset(np.zeros(
(imSize[0] * row_range, patchSize**2)),
borrow=True)
classify = theano.function([],
self.p_y_given_x,
givens={self.x: data_shared})
for row in xrange(0, 1024, row_range):
if row % 100 == 0:
print row
#data = Utility.get_patch(image, row, row_range, patchSize )
#print 'data shape:', data.shape
'''
data = Utility.generate_patch_data_rows(
image,
rowOffset=row,
rowRange=row_range,
patchSize=patchSize,
imSize=imSizePadded,
data_mean=mean,
data_std=std)
'''
data = Utility.get_patch(image,
row,
row_range,
patchSize,
data_mean=mean,
data_std=std)
#print 'data:', np.shape(data)
data_shared.set_value(np.float32(data))
result = classify()
#print 'results:', np.shape(result)
membraneProbabilities[row, :] = result[:, 0]
end_time = time.clock()
total_time = (end_time - start_time)
print >> sys.stderr, ('Running time: ' + '%.2fm' % (total_time / 60.))
return np.array(membraneProbabilities)