def res_layer(inp, chl):
pre = inp
#inp = conv_bn(inp, 3, 1, 1, chl, True)
#inp = conv_bn(inp, 3, 1, 1, chl, False)
inp = den_lay(inp, chl)
inp = arith.ReLU(inp)
inp = den_lay(inp, chl)
inp = arith.ReLU(inp + pre)
return inp
def relu_conv_bn(inp,
ker_shape,
stride,
padding,
out_chl,
isrelu=True,
isbn=True):
global idx
idx += 1
if isrelu:
inp = arith.ReLU(inp)
inp = Conv2D("conv{}".format(idx),
inp,
kernel_shape=ker_shape,
stride=stride,
padding=padding,
output_nr_channel=out_chl,
nonlinearity=Identity())
if isbn:
inp = BN("bn{}".format(idx), inp, eps=1e-9)
inp = ElementwiseAffine("bnaff{}".format(idx),
inp,
shared_in_channels=False,
k=C(1),
b=C(0))
return inp
def bn_relu_conv(inp, ker_shape, stride, padding, out_chl, has_relu, has_bn, has_conv = True):
global idx
idx += 1
if has_bn:
l1 = BN("bn{}".format(idx), inp, eps = 1e-9)
l1 = ElementwiseAffine("bnaff{}".format(idx), l1, shared_in_channels = False, k = C(1), b = C(0))
else:
l1 = inp
if has_relu:
l2 = arith.ReLU(l1)
else:
l2 = l1
if not has_conv:
return l2
l3 = Conv2D(
"conv{}".format(idx), l2, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
#W = G(mean = 0, std = (1 / (ker_shape**2 * inp.partial_shape[1]))**0.5),
#b = C(0),
nonlinearity = Identity()
)
return l3
def conv_bn(inp, ker_shape, stride, padding, out_chl, isrelu, mode = None):
global idx
idx += 1
print(inp.partial_shape, ker_shape, out_chl)
if ker_shape == 1:
W = ortho_group.rvs(out_chl)
W = W[:, :inp.partial_shape[1]]
W = W.reshape(W.shape[0], W.shape[1], 1, 1)
W = ConstProvider(W)
b = ConstProvider(np.zeros(out_chl))
else:
W = G(mean = 0, std = ((1 + int(isrelu)) / (ker_shape**2 * inp.partial_shape[1]))**0.5)
b = C(0)
l1 = Conv2D(
"conv{}".format(idx), inp, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
group = mode,
W = W,
b = b,
nonlinearity = Identity()
)
l2 = BN("bn{}".format(idx), l1, eps = 1e-9)
l2 = ElementwiseAffine("bnaff{}".format(idx), l2, shared_in_channels = False, k = C(1), b = C(0))
if isrelu:
l2 = arith.ReLU(l2)
return l2, l1
def res_layer(inp, chl, group, stride=1, proj=False, shift=None):
pre = inp
if group == 1:
inp = conv_bn(inp, 1, stride, 0, chl // 4, True, group=group)
inp = conv_bn(inp, 3, 1, 1, chl // 4, True)
inp = conv_bn(inp, 1, 1, 0, chl, False, group=group)
else:
"""
lay1 = conv_bn(inp, 3, 1, 1, chl // 4, True, group = group)
inp = O.Concat([inp[:, shift * chl // 4 // group:, :, :], inp[:, :shift * chl // 4 // group, :, :]], axis = 1)
lay2 = conv_bn(inp, 3, 1, 1, chl // 4, True, group = group)
inp = lay1 + lay2
"""
inp = conv_bn(inp, 1, stride, 0, chl // 4, True, group=group)
"""
subchl = chl // 4 // group
inp = inp.reshape(inp.shape[0], group, subchl, inp.shape[2], inp.shape[3])
inp = inp.dimshuffle(0, 2, 1, 3, 4)
inp = inp.reshape(inp.shape[0], chl // 4, inp.shape[3], inp.shape[4])
"""
inp = conv_bn(inp, 3, 1, 1, chl // 4, True, group=group)
inp = conv_bn(inp, 1, 1, 0, chl, False, group=group)
if proj:
pre = conv_bn(pre, 1, stride, 0, chl, False, group=group)
inp = arith.ReLU(inp + pre)
return inp
def res_layer(inp, chl, stride = 1, proj = False, se = None):
pre = inp
inp = conv_bn(inp, 1, stride, 0, chl // 4, True)
inp = conv_bn(inp, 3, 1, 1, chl // 4, True)
inp = conv_bn(inp, 1, 1, 0, chl, False)
if proj:
pre = conv_bn(pre, 1, stride, 0, chl, False)
name = inp.name
#Global Average Pooling
SE = inp.mean(axis = 3).mean(axis = 2)
#fc0
SE = FullyConnected(
"fc0({})".format(name), SE, output_dim = chl // 4,
nonlinearity = ReLU()
)
#fc1
if se is None:
se = SE
else:
se = O.Concat([se, SE], axis = 1)
SE = FullyConnected(
"fc1({})".format(name), se, output_dim = chl,
nonlinearity = Sigmoid()
)
se = FullyConnected(
"fc({})".format(se.name), se, output_dim = chl // 4,
nonlinearity = ReLU()
)
inp = inp * SE.dimshuffle(0, 1, 'x', 'x')
inp = arith.ReLU(inp + pre)
return inp, se
def res_layer(inp, chl, group, stride=1, proj=False, shift=None):
pre = inp
if group == 1:
inp = conv_bn(inp, 1, stride, 0, chl // 4, True, group=group)
inp = conv_bn(inp, 3, 1, 1, chl // 4, True)
inp = conv_bn(inp, 1, 1, 0, chl, False, group=group)
else:
"""
lay1 = conv_bn(inp, 3, 1, 1, chl // 4, True, group = group)
inp = O.Concat([inp[:, shift * chl // 4 // group:, :, :], inp[:, :shift * chl // 4 // group, :, :]], axis = 1)
lay2 = conv_bn(inp, 3, 1, 1, chl // 4, True, group = group)
inp = lay1 + lay2
"""
inp = conv_bn(inp,
1,
stride,
0,
chl // 4,
True,
group=group,
shift=shift)
inp = conv_bn(inp, 3, 1, 1, chl // 4, True, group=group, shift=shift)
inp = conv_bn(inp, 1, 1, 0, chl, False, group=group, shift=shift)
if proj:
pre = conv_bn(pre, 1, stride, 0, chl, False, group=group)
inp = arith.ReLU(inp + pre)
return inp
def res_layer(inp, chl, stride = 1, proj = False):
pre = inp
inp = conv_bn(inp, 1, stride, 0, chl // 4, True)
#inp = conv_bn(inp, 3, 1, 1, chl // 4, True)
inp = den_layer(inp, chl // 4)
inp = conv_bn(inp, 1, 1, 0, chl, False)
if proj:
pre = conv_bn(pre, 1, stride, 0, chl, False)
name = inp.name
#Global Average Pooling
SE = inp.mean(axis = 3).mean(axis = 2)
sum_lay = 0
out_lay = 0
lay = FullyConnected(
"fc0({})".format(name), SE, output_dim = chl,
nonlinearity = ReLU()
)
#fc1
lay = FullyConnected(
"fc1({})".format(name), lay, output_dim = chl,
nonlinearity = Sigmoid()
)
inp = inp * lay.dimshuffle(0, 1, 'x', 'x')
inp = arith.ReLU(inp + pre)
return inp
def bn_relu_conv(inp, ker_shape, stride, padding, out_chl, has_relu, has_bn, has_conv = True):
global idx
idx += 1
if has_bn:
l1 = BN("bn{}".format(idx), inp, eps = 1e-9)
l1 = ElementwiseAffine("bnaff{}".format(idx), l1, shared_in_channels = False, k = C(1), b = C(0))
else:
l1 = inp
if has_relu:
l2 = arith.ReLU(l1)
else:
l2 = l1
if not has_conv:
return l2, None
l3 = Conv2D(
"conv{}".format(idx), l2, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
nonlinearity = Identity()
)
w = l3.inputs[1]
assert ":W" in w.name
return l3, w
def res_layer(inp, chl, stride=1, proj=False):
pre = inp
inp = conv_bn(inp, 1, stride, 0, chl // 4, True)
name = inp.name
#Global Average Pooling
SE = inp.mean(axis=3).mean(axis=2)
sum_lay = 0
out_lay = 0
width = 4
lay = FullyConnected("fc0({})".format(name),
SE,
output_dim=chl // 4,
nonlinearity=ReLU())
#fc1
lay = FullyConnected("fc1({})".format(name),
lay,
output_dim=chl // 4 * width,
nonlinearity=Identity())
lay = lay.reshape(inp.shape[0], chl // 4, width)
lay = Softmax("softmax({})".format(name), lay, axis=2)
for i in range(width):
if i == 0:
inp_lay = inp
else:
inp_lay = O.Concat([inp[:, width:, :, :], inp[:, :width, :, :]],
axis=1)
inp_lay = inp_lay * lay[:, :, i].dimshuffle(0, 1, 'x', 'x')
inp = O.ReLU(inp_lay)
inp = conv_bn(inp, 3, 1, 1, chl // 4, True)
inp = conv_bn(inp, 1, 1, 0, chl, False)
if proj:
pre = conv_bn(pre, 1, stride, 0, chl, False)
inp = arith.ReLU(inp + pre)
return inp
def res_layer(inp, chl, stride=1, proj=False):
pre = inp
inp = conv_bn(inp, 1, stride, 0, chl // 4, True)
inp = conv_bn(inp, 3, 1, 1, chl // 4, True)
inp = conv_bn(inp, 1, 1, 0, chl, False)
if proj:
pre = conv_bn(pre, 1, stride, 0, chl, False)
inp = arith.ReLU(inp + pre)
return inp
def den_lay(inp, chl):
out = []
stage = 8
for i in range(stage):
lay = conv_bn(inp, 3, 1, 1, chl // stage, False)
out.append(lay)
lay = arith.ReLU(lay)
inp = O.Concat([inp, lay], axis=1)
return O.Concat(out, axis=1)
def deconv_bn_relu(name, inp, kernel_shape = None, stride = None, padding = None, output_nr_channel = None, isbnrelu = True): lay = O.Deconv2DVanilla(name, inp, kernel_shape = kernel_shape, stride = stride, padding = padding, output_nr_channel = output_nr_channel) if isbnrelu: lay = BN(name + "bn", lay, eps = 1e-9) lay = ElementwiseAffine(name + "bnaff", lay, shared_in_channels = False, k = C(1), b = C(0)) lay = arith.ReLU(lay) return lay
def conv_bn(inp, ker_shape, stride, padding, out_chl, isrelu):
global idx
idx += 1
l1 = Conv2D(
"conv{}".format(idx), inp, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
nonlinearity = Identity()
)
l2 = BN("bn{}".format(idx), l1, eps = 1e-9)
l2 = ElementwiseAffine("bnaff{}".format(idx), l2, shared_in_channels = False, k = C(1), b = C(0))
if isrelu:
l2 = arith.ReLU(l2)
return l2
def res_block(inp, chl, n):
stride = 2
if chl == 16:
stride = 1
pre = inp
inp = bn_relu_conv(inp, 3, stride, 1, chl, True, True)
inp = bn_relu_conv(inp, 3, 1, 1, chl, True, True)
inp = inp + bn_relu_conv(pre, 1, stride, 0, chl, True, True)
inp = arith.ReLU(inp)
for i in range(n - 1):
inp = res_layer(inp, chl)
return inp
def res_block(inp, chl, n): stride = 2 if chl == 16: stride = 1 pre = inp inp = conv_bn(inp, 3, stride, 1, chl, True) inp = conv_bn(inp, 3, 1, 1, chl, False) inp = inp + conv_bn(pre, 1, stride, 0, chl, False) inp = arith.ReLU(inp) for i in range(n - 1): inp = res_layer(inp, chl) return inp
def conv_bn(inp, ker_shape, stride, padding, out_chl, isrelu, group = 1, shift = 0):
global idx
idx += 1
if group == 1:
l1 = Conv2D(
"conv{}".format(idx), inp, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
#W = G(mean = 0, std = ((1) / (ker_shape**2 * inp.partial_shape[1]))**0.5),
#b = C(0),
nonlinearity = Identity()
)
else:
if shift == 0:
l1 = Conv2D(
"conv{}".format(idx), inp, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
#W = G(mean = 0, std = ((1) / (ker_shape**2 * inp.partial_shape[1]))**0.5),
#b = C(0),
nonlinearity = Identity(),
group = group,
)
else:
shift = 1
l1 = inp
while shift != group:
l11 = Conv2D(
"conv{}_{}_1".format(idx, shift), l1, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
#W = G(mean = 0, std = ((1) / (ker_shape**2 * inp.partial_shape[1]))**0.5),
#b = C(0),
nonlinearity = Identity(),
group = group,
)
inp_chl = l1.partial_shape[1]
l1 = O.Concat([l1[:, shift * inp_chl // group:, :, :], l1[:, :shift * inp_chl // group, :, :]], axis = 1)
l12 = Conv2D(
"conv{}_{}_2".format(idx, shift), l1, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
#W = G(mean = 0, std = ((1) / (ker_shape**2 * inp.partial_shape[1]))**0.5),
#b = C(0),
nonlinearity = Identity(),
group = group,
)
l1 = l11 + l12
shift *= 2
l2 = BN("bn{}".format(idx), l1, eps = 1e-9)
l2 = ElementwiseAffine("bnaff{}".format(idx), l2, shared_in_channels = False, k = C(1), b = C(0))
if isrelu:
l2 = arith.ReLU(l2)
return l2
def conv_bn(inp, ker_shape, stride, padding, out_chl, isrelu):
global idx
idx += 1
l1 = Conv2D(
"encoder_conv{}".format(idx), inp, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
W = G(mean = 0, std = ((1 + int(isrelu)) / (ker_shape**2 * inp.partial_shape[1]))**0.5),
nonlinearity = Identity()
)
l2 = BN("encoder_bn{}".format(idx), l1, eps = 1e-9)
l2 = ElementwiseAffine("bnaff{}".format(idx), l2, shared_in_channels = False, k = C(1), b = C(0))
if isrelu:
l2 = arith.ReLU(l2)
return l2, l1
def res_layer(inp, chl, stride=1, proj=False):
pre = inp
inp = conv_bn(inp, 1, stride, 0, chl // 4, True)
inp = conv_bn(inp, 3, 1, 1, chl // 4, True)
inp = conv_bn(inp, 1, 1, 0, chl, False)
name = inp.name
inp = ElementwiseAffine("aff({})".format(name),
inp,
shared_in_channels=False,
k=C(0.5),
b=C(0))
if proj:
pre = conv_bn(pre, 1, stride, 0, chl, False)
inp = arith.ReLU(inp + pre)
return inp
def bn_relu_conv(inp, ker_shape, stride, padding, out_chl, isrelu, isbn):
global idx
idx += 1
if isbn:
inp = BN("bn{}".format(idx), inp, eps = 1e-9)
inp = ElementwiseAffine("bnaff{}".format(idx), inp, shared_in_channels = False, k = C(1), b = C(0))
if isrelu:
inp = arith.ReLU(inp)
inp = Conv2D(
"conv{}".format(idx), inp, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
#W = G(mean = 0, std = ((1) / (ker_shape**2 * inp.partial_shape[1]))**0.5),
#b = C(0),
nonlinearity = Identity()
)
return inp
def conv_wn(inp, ker_shape, stride, padding, out_chl, isrelu):
global idx
idx += 1
l1 = Conv2D(
"conv{}".format(idx), inp, kernel_shape = ker_shape, stride = stride, padding = padding,
output_nr_channel = out_chl,
W = G(mean = 0, std = 0.05),
nonlinearity = Identity()
)
W = l1.inputs[1]
#l2 = BN("bn{}".format(idx), l1, eps = 1e-9)
w = l1.inputs[1]
assert ":W" in w.name
w = (w**2).sum(axis = 3).sum(axis = 2).sum(axis = 1)**0.5
l1 = l1 / w.dimshuffle('x', 0, 'x', 'x')
l2 = ElementwiseAffine("bnaff{}".format(idx), l1, shared_in_channels = False, k = C(1), b = C(0))
if isrelu:
l2 = arith.ReLU(l2)
return l2, l1, W
def res_block(inp, chl, n): lis_w = [] stride = 2 if chl == 16: stride = 1 pre = inp inp, w = conv_bn(inp, 3, stride, 1, chl, True) lis_w.append(w) inp, w = conv_bn(inp, 3, 1, 1, chl, False) lis_w.append(w) res_path, w = conv_bn(pre, 1, stride, 0, chl, False) inp = inp + res_path lis_w.append(w) inp = arith.ReLU(inp) for i in range(n - 1): inp, lis_new = res_layer(inp, chl) lis_w += lis_new return inp, lis_w
def res_layer(inp, chl):
pre = inp
inp = conv_bn(inp, 3, 1, 1, chl, True)
inp = conv_bn(inp, 3, 1, 1, chl, False)
name = inp.name
#Global Average Pooling
SE = inp.mean(axis=3).mean(axis=2)
#fc0
SE = FullyConnected("fc0({})".format(name),
SE,
output_dim=SE.partial_shape[1],
nonlinearity=ReLU())
#fc1
SE = FullyConnected("fc1({})".format(name),
SE,
output_dim=SE.partial_shape[1],
nonlinearity=Sigmoid())
inp = inp * SE.dimshuffle(0, 1, 'x', 'x')
inp = arith.ReLU(inp + pre)
return inp
def res_layer(inp, chl):
pre = inp
inp = conv_bn(inp, 3, 1, 1, chl, True)
inp = conv_bn(inp, 3, 1, 1, chl, False)
name = inp.name
#Global Average Pooling
SE = inp.mean(axis=3).mean(axis=2)
group = 1
#fc0
SE = FullyConnected("fc0({})".format(name),
SE,
output_dim=chl,
nonlinearity=ReLU())
#fc1
SE = FullyConnected("fc1({})".format(name),
SE,
output_dim=(chl // group)**2 * group,
nonlinearity=Sigmoid())
SE = SE.reshape(inp.shape[0] * group, chl // group, chl // group, 1, 1)
w = SE
SE /= SE.sum(axis=4).sum(axis=3).sum(axis=2).dimshuffle(
0, 1, "x", "x", "x")
#inp = inp * SE.dimshuffle(0, 1, 'x', 'x')
inp = inp.reshape(1, inp.shape[0] * inp.shape[1], inp.shape[2],
inp.shape[3])
inp = Conv2D(
"conv({})".format(name),
inp,
kernel_shape=1,
stride=1,
padding=0,
#output_nr_channel = chl,
W=SE,
nonlinearity=Identity(),
#group = group
)
inp = inp.reshape(pre.shape)
inp = arith.ReLU(inp + pre)
return inp, w
def conv_bn(inp, ker_shape, stride, padding, out_chl, isrelu):
global idx
idx += 1
l10 = Conv2D("conv{}_0".format(idx),
inp,
kernel_shape=ker_shape,
stride=stride,
padding=padding,
output_nr_channel=out_chl // 2,
W=G(mean=0,
std=((1 + int(isrelu)) /
(ker_shape**2 * inp.partial_shape[1]))**0.5),
nonlinearity=Identity())
l11 = Conv2D("conv{}_1".format(idx),
inp,
kernel_shape=ker_shape,
stride=stride,
padding=padding,
output_nr_channel=out_chl // 2,
W=G(mean=0,
std=((1 + int(isrelu)) /
(ker_shape**2 * inp.partial_shape[1]))**0.5),
nonlinearity=Identity())
W = l11.inputs[1].owner_opr
b = l11.inputs[2].owner_opr
W.set_freezed()
b.set_freezed()
l1 = Concat([l10, l11], axis=1)
l2 = BN("bn{}".format(idx), l1, eps=1e-9)
l2 = ElementwiseAffine("bnaff{}".format(idx),
l2,
shared_in_channels=False,
k=C(1),
b=C(0))
if isrelu:
l2 = arith.ReLU(l2)
return l2, l1
def dfconv(inp, chl, isrelu, ker_shape = 3, stride = 1, padding = 1, dx = [-1, 0, 1], dy = [-1, 0, 1]):
global idx
name = "conv{}".format(idx)
offsetlay = Conv2D(
name + "conv1", inp, kernel_shape = 3, stride = 1, padding = 1,
output_nr_channel = ker_shape**2,
W = G(mean = 0, std = ((1) / (3**2 * inp.partial_shape[1]))**0.5),
nonlinearity = Identity()
)
offsetlay = BN(name + "BN1", offsetlay, eps = 1e-9)
offsetlay = arith.ReLU(offsetlay)
offsetlay = Conv2D(
name + "conv2", inp, kernel_shape = 3, stride = 1, padding = 1,
output_nr_channel = ker_shape**2,
W = G(mean = 0, std = ((1) / (3**2 * inp.partial_shape[1]))**0.5),
nonlinearity = Identity()
)
offsetlay = BN(name + "BN2", offsetlay, eps = 1e-9)
offsetx = inp.partial_shape[2] * Conv2D(
name + "offsetx", offsetlay, kernel_shape = ker_shape, stride = stride,
padding = padding,
output_nr_channel = ker_shape**2,
W = G(mean = 0, std = (1 / (ker_shape**2 * inp.partial_shape[2]))**0.5),
nonlinearity = Identity()
)
offsety = inp.partial_shape[3] * Conv2D(
name + "offsety", offsetlay, kernel_shape = ker_shape, stride = stride,
padding = padding,
output_nr_channel = ker_shape**2,
W = G(mean = 0, std = (1 / (ker_shape**2 * inp.partial_shape[3]))**0.5),
nonlinearity = Identity()
)
"""
gamma = 0.0001
ndim = ker_shape**2 * offsetx.partial_shape[2] * offsetx.partial_shape[3]
offsetx = FullyConnected(
name + "offsetx", offsetx, output_dim = ndim,
W = G(mean = 0, std = (1 / ndim)**0.5),
b = C(0),
nonlinearity = Identity()
)
offsetx = offsetx.reshape(offsety.shape)
offsety = FullyConnected(
name + "offsety", offsety, output_dim = ndim,
W = G(mean = 0, std = (1 / ndim)**0.5),
b = C(0),
nonlinearity = Identity()
)
offsety = offsety.reshape(offsetx.shape)
"""
outputs = []
for sx in range(2):
for sy in range(2):
if sx == 0:
ofx = Floor(offsetx)
bilx = offsetx - ofx
else:
ofx = Ceil(offsetx)
bilx = ofx - offsetx
if sy == 0:
ofy = Floor(offsety)
bily = offsety - ofy
else:
ofy = Ceil(offsety)
bily = ofy - offsety
"""
No padding
padding1 = ConstProvider(np.zeros((inp.partial_shape[0], inp.partial_shape[1], 1, inp.partial_shape[3])))
padding2 = ConstProvider(np.zeros((inp.partial_shape[0], inp.partial_shape[1], inp.partial_shape[2] + 2, 1)))
arg_fea = Concat([padding1, inp, padding1], axis = 2)
arg_fea = Concat([padding2, arg_fea, padding2], axis = 3)
"""
arg_fea = inp
#one_mat = ConstProvider(np.ones((inp.partial_shape[2], inp.partial_shape[3])), dtype = np.int32)
one_mat = ConstProvider(1, dtype = np.int32).add_axis(0).broadcast((ofx.partial_shape[2], ofx.partial_shape[3]))
affx = (Cumsum(one_mat, axis = 0) - 1) * stride
affy = (Cumsum(one_mat, axis = 1) - 1) * stride
ofx = ofx + affx.dimshuffle('x', 'x', 0, 1)
ofy = ofy + affy.dimshuffle('x', 'x', 0, 1)
one_mat = ConstProvider(np.ones((ker_shape, ofx.partial_shape[2], ofx.partial_shape[3])))
#ofx[:, :ker_shape, :, :] -= 1
#ofx[:, ker_shape*2:, :, :] += 1
ofx += Concat([one_mat * i for i in dx], axis = 0).dimshuffle('x', 0, 1, 2)
#ofy[:, ::3, :, :] -= 1
#ofy[:, 2::3, :, :] += 1
one_mat = ones((1, ofx.partial_shape[2], ofx.partial_shape[3]))
one_mat = Concat([one_mat * i for i in dy], axis = 0)
one_mat = Concat([one_mat] * ker_shape, axis = 0)
ofy += one_mat.dimshuffle('x', 0, 1, 2)
ofx = Max(Min(ofx, arg_fea.partial_shape[2] - 1), 0)
ofy = Max(Min(ofy, arg_fea.partial_shape[3] - 1), 0)
def DeformReshape(inp, ker_shape):
inp = inp.reshape(inp.shape[0], ker_shape, ker_shape, inp.shape[2], inp.shape[3])
inp = inp.dimshuffle(0, 3, 1, 4, 2)
inp = inp.reshape(inp.shape[0], inp.shape[1] * inp.shape[2], inp.shape[3] * inp.shape[4])
return inp
ofx = DeformReshape(ofx, ker_shape)
ofy = DeformReshape(ofy, ker_shape)
bilx = DeformReshape(bilx, ker_shape)
bily = DeformReshape(bily, ker_shape)
of = ofx * arg_fea.shape[2] + ofy
arg_fea = arg_fea.reshape(arg_fea.shape[0], arg_fea.shape[1], -1)
of = of.reshape(ofx.shape[0], -1)
of = of.dimshuffle(0, 'x', 1)
#of = Concat([of] * arg_fea.partial_shape[1], axis = 1)
of = of.broadcast((of.shape[0], arg_fea.shape[1], of.shape[2]))
arx = Linspace(0, arg_fea.shape[0], arg_fea.shape[0], endpoint = False)
arx = arx.add_axis(1).add_axis(2).broadcast(of.shape)
ary = Linspace(0, arg_fea.shape[1], arg_fea.shape[1], endpoint = False)
ary = ary.add_axis(0).add_axis(2).broadcast(of.shape)
of = of.add_axis(3)
arx = arx.add_axis(3)
ary = ary.add_axis(3)
idxmap = Astype(Concat([arx, ary, of], axis = 3), np.int32)
"""
sample = []
for i in range(arg_fea.partial_shape[0]):
for j in range(arg_fea.partial_shape[1]):
sample.append(arg_fea[i][j].ai[of[i][j]].dimshuffle('x', 0))
sample = Concat(sample, axis = 0)
"""
sample = IndexingRemap(arg_fea, idxmap).reshape(inp.shape[0], inp.shape[1], bilx.shape[1], -1)
bilx = bilx.dimshuffle(0, 'x', 1, 2).broadcast(sample.shape)
bily = bily.dimshuffle(0, 'x', 1, 2).broadcast(sample.shape)
sample *= bilx * bily
outputs.append(sample)
output = outputs[0]
for i in outputs[1:]:
output += i
return conv_bn(output, ker_shape, 3, 0, chl, isrelu)
def res_layer(inp, chl): pre = inp inp, w = conv_bn(inp, 3, 1, 1, chl, True) inp, w = conv_bn(inp, 3, 1, 1, chl, False) inp = arith.ReLU(inp + pre) return inp, [w, w]