def test(self, input_matrix, teacher_matrix):
"""Function to test the network
@param input_matrix -- matrix consisting of input
data to the network.
@param teacher_matrix -- matrix consisting of labels
of input data .
"""
number_of_pictures = input_matrix.shape[-1]
mse = 0
squared_errors = cp.dev_matrix_cmf(self.neuron_layer[-1].deltas.h,
self.neuron_layer[-1].deltas.w)
for batch in xrange(number_of_pictures/self.batch_size):
index_begin = self.batch_size * batch
index_end = index_begin + self.batch_size
self.neuron_layer[0].activations = cp.push( input_matrix[:,
index_begin:index_end].astype('float32').copy('F'))
teachbatch = cp.push(teacher_matrix[:,
index_begin:index_end].astype('float32').copy('F'))
for i in xrange(self.number_of_layers):
self.weight_layer[i].forward()
cp.apply_binary_functor(squared_errors, self.neuron_layer[-1].deltas,
cp.binary_functor.COPY)
cp.apply_scalar_functor(squared_errors, cp.scalar_functor.SQUARE)
mse += cp.sum(squared_errors)
teachbatch.dealloc()
print "MSE: ", (mse/number_of_pictures)
squared_errors.dealloc()
def loadLastLayer(self,dim1,dim2):
fn=os.path.join(self.cfg.workdir,"weights-%d-finetune.npy"%(self.NumberOfLayers-1))
fn_bias=os.path.join(self.cfg.workdir,"bias-%d-finetune.npy"%(self.NumberOfLayers-1))
if os.path.exists(fn) and os.path.exists(fn_bias):
top_weights=np.load(fn)
assert((dim1,dim2)==top_weights.shape)
self.Weights.append(cp.push(top_weights))
top_bias=np.load(fn_bias)
assert(dim2==top_bias.shape[0])
self.Bias.append(cp.push(top_bias))
return 1
else:
return 0
def loadLastLayer(self, dim1, dim2):
fn = os.path.join(
self.cfg.workdir,
"weights-%d-finetune.npy" % (self.NumberOfLayers - 1))
fn_bias = os.path.join(
self.cfg.workdir,
"bias-%d-finetune.npy" % (self.NumberOfLayers - 1))
if os.path.exists(fn) and os.path.exists(fn_bias):
top_weights = np.load(fn)
assert ((dim1, dim2) == top_weights.shape)
self.Weights.append(cp.push(top_weights))
top_bias = np.load(fn_bias)
assert (dim2 == top_bias.shape[0])
self.Bias.append(cp.push(top_bias))
return 1
else:
return 0
def gray_test(ni):
src = cp.push(to_cmuc(np.tile(ni,(1,4))))
dst = cp.dev_matrix_cmf(src.h,src.w)
cp.fill(dst,0)
cp.image_move(dst,src,128,128,1,-10,-4)
res = cp.pull(dst)
#set_trace()
plt.matshow(res[0:128**2,0].reshape(128,128))
plt.colorbar()
plt.show()
def gray_test(ni):
src = cp.push(to_cmuc(np.tile(ni, (1, 4))))
dst = cp.dev_matrix_cmf(src.h, src.w)
cp.fill(dst, 0)
cp.image_move(dst, src, 128, 128, 1, -10, -4)
res = cp.pull(dst)
#set_trace()
plt.matshow(res[0:128**2, 0].reshape(128, 128))
plt.colorbar()
plt.show()
def color_test(ni):
ts = 128
src = cp.push(to_cmuc(np.tile(ni,(1,4))))
dst = cp.dev_matrix_cmf(ts**2*3,src.w)
cp.fill(dst,0)
cp.image_move(dst,src,128,ts,4,-10,-4)
res = cp.pull(dst)
plt.matshow(res[0:ts**2,0].reshape(ts,ts), cmap = plt.cm.bone_r)
plt.matshow(res[ts**2:2*ts**2,0].reshape(ts,ts), cmap = plt.cm.bone_r)
plt.matshow(res[2*ts**2:3*ts**2,0].reshape(ts,ts), cmap = plt.cm.bone_r)
plt.show()
def color_test(ni):
ts = 128
src = cp.push(to_cmuc(np.tile(ni, (1, 4))))
dst = cp.dev_matrix_cmf(ts**2 * 3, src.w)
cp.fill(dst, 0)
cp.image_move(dst, src, 128, ts, 4, -10, -4)
res = cp.pull(dst)
plt.matshow(res[0:ts**2, 0].reshape(ts, ts), cmap=plt.cm.bone_r)
plt.matshow(res[ts**2:2 * ts**2, 0].reshape(ts, ts), cmap=plt.cm.bone_r)
plt.matshow(res[2 * ts**2:3 * ts**2, 0].reshape(ts, ts),
cmap=plt.cm.bone_r)
plt.show()
def test_pixel_classes():
w, h = 512, 512
input_channels, pyramid_channels = 4, 3
pic = Image.open("tests/data/lena.bmp").resize((w, h)).convert("RGBA")
pic = np.asarray(pic).reshape(h, w * 4)
pic_d = cp.push(pic)
pyr = cp.dev_image_pyramid_f(pic_d.h / 2, pic_d.w / input_channels / 2, 4,
pyramid_channels)
pyr.build(pic_d, 4)
plt.matshow(pic[0:h:2, 0:4 * w:8])
#plt.matshow(cp.pull(pyr.get(1,0)))
#plt.title("Channel0")
#plt.matshow(cp.pull(pyr.get(1,1)))
#plt.title("Channel1")
#plt.matshow(cp.pull(pyr.get(1,2)))
#plt.title("Channel2")
#plt.matshow(cp.pull(pyr.get_all_channels(1)))
#plt.title("allchannels level 1")
#plt.show()
# create source image from higher level of pyramid
pic1 = pyr.get_all_channels(0)
for i in xrange(10):
smooth(pic1)
plt.matshow(cp.pull(pic1)[:h / 2, :w])
ca = cp.dev_cuda_array_f(pic1.h, pic1.w, 1)
ca.assign(pic1)
# create destination matrix
pic0 = pyr.get(0)
dst = cp.dev_matrix_rmuc(pic0.h, pic0.w * 4) # uchar4
# generate pixel classes and visualize
cp.get_pixel_classes(dst, ca, 1)
tmp = cp.pull(dst)
tmp = Image.frombuffer("CMYK", (pic0.w, pic0.h),
cp.pull(dst).flatten(), "raw", "CMYK", 0, 1).resize(
(2 * 512, 2 * 512), Image.NEAREST)
tmp.show()
print cp.pull(dst)
plt.show()
def test_pixel_classes():
w, h = 512,512
input_channels, pyramid_channels = 4,3
pic = Image.open("tests/data/lena.bmp").resize((w,h)).convert("RGBA")
pic = np.asarray(pic).reshape(h,w*4)
pic_d = cp.push(pic)
pyr = cp.dev_image_pyramid_f(pic_d.h/2,pic_d.w/input_channels/2,4,pyramid_channels)
pyr.build(pic_d,4)
plt.matshow(pic[0:h:2,0:4*w:8])
#plt.matshow(cp.pull(pyr.get(1,0)))
#plt.title("Channel0")
#plt.matshow(cp.pull(pyr.get(1,1)))
#plt.title("Channel1")
#plt.matshow(cp.pull(pyr.get(1,2)))
#plt.title("Channel2")
#plt.matshow(cp.pull(pyr.get_all_channels(1)))
#plt.title("allchannels level 1")
#plt.show()
# create source image from higher level of pyramid
pic1 = pyr.get_all_channels(0)
for i in xrange(10): smooth(pic1)
plt.matshow(cp.pull(pic1)[:h/2,:w])
ca = cp.dev_cuda_array_f(pic1.h,pic1.w,1)
ca.assign(pic1)
# create destination matrix
pic0 = pyr.get(0)
dst = cp.dev_matrix_rmuc(pic0.h,pic0.w*4) # uchar4
# generate pixel classes and visualize
cp.get_pixel_classes(dst,ca,1)
tmp = cp.pull(dst)
tmp = Image.frombuffer("CMYK", (pic0.w,pic0.h), cp.pull(dst).flatten(), "raw", "CMYK", 0, 1 ).resize((2*512,2*512), Image.NEAREST)
tmp.show()
print cp.pull(dst)
plt.show()
def build_pyramid_GPU(pic,input_channels,pyramid_channels):
pic_d = cp.push(pic)
pyr = cp.dev_image_pyramid_f(pic_d.h/2,pic_d.w/input_channels/2,4,pyramid_channels)
pyr.build(pic_d,4)
def build_pyramid_GPU(pic, input_channels, pyramid_channels):
pic_d = cp.push(pic)
pyr = cp.dev_image_pyramid_f(pic_d.h / 2, pic_d.w / input_channels / 2, 4,
pyramid_channels)
pyr.build(pic_d, 4)
class MLP:
"""
A Multi-Layer Perceptron
"""
def __init__(self, neurons, batch_size):
"""Constructor
@param neurons -- array of sizes of layers.
@param batch_size -- size of batch being used for training.
"""
self.number_of_layers = len(neurons) - 1
self.batch_size = batch_size
self.neuron_layer = []
self.weight_layer = []
for i in xrange(self.number_of_layers+1):
dim1 = neurons[i]
self.neuron_layer.append(neuron_layer(dim1,
self.batch_size ))
for i in xrange(self.number_of_layers):
self.weight_layer.append(weight_layer(self.neuron_layer[i],
self.neuron_layer[i+1]))
def train(self, input_matrix, teacher_matrix, number_of_epochs):
"""Function to train the network
@param input_matrix -- matrix consisting of input data
to the network.
@param teacher_matrix -- matrix consisting of labels
of input data.
@param number_of_epochs -- number of rounds the network
is to be trained.
"""
number_of_pictures = input_matrix.shape[-1]
squared_errors = cp.dev_matrix_cmf(self.neuron_layer[-1].deltas.h,
self.neuron_layer[-1].deltas.w)
for r in xrange(number_of_epochs):
print "Epoch ", r+1, "/", number_of_epochs
mse = 0
for batch in xrange(number_of_pictures/self.batch_size):
index_begin = self.batch_size * batch
index_end = self.batch_size + index_begin
# Push input and teacher to GPU memory
self.neuron_layer[0].activations = cp.push(
input_matrix[:,index_begin:index_end].astype('float32').copy('F'))
teachbatch = cp.push(
teacher_matrix[:,index_begin:index_end].astype('float32').copy('F'))
# Forward-Pass
for i in xrange(self.number_of_layers):
self.weight_layer[i].forward()
# calculate error at output layer
cp.apply_binary_functor(self.neuron_layer[-1].deltas,
teachbatch, cp.binary_functor.COPY)
cp.apply_binary_functor(self.neuron_layer[-1].deltas,
self.neuron_layer[-1].activations,
cp.binary_functor.SUBTRACT)
cp.apply_binary_functor(squared_errors, self.neuron_layer[-1].deltas,
cp.binary_functor.COPY)
cp.apply_scalar_functor(squared_errors, cp.scalar_functor.SQUARE)
mse += cp.sum(squared_errors)
# Backward-Pass
for i in xrange(self.number_of_layers):
self.weight_layer[self.number_of_layers-i-1].backward()
# Don't wait for garbage collector
teachbatch.dealloc()
self.neuron_layer[0].activations.dealloc()
print "MSE: ", (mse/number_of_pictures)
squared_errors.dealloc()
