def _concat_new_axis(t1, t2, t3, t4, axis=1):
return Concat(
[t1.add_axis(-1),
t2.add_axis(-1),
t3.add_axis(-1),
t4.add_axis(-1)],
axis=axis)
def dense_block(inp, k, l):
lay = inp
for i in range(l):
cur_lay = bn_relu_conv(lay, 3, 1, 1, k, True, True)
name = cur_lay.name
group = k // 4
#G.P.
SE = cur_lay.mean(axis=3).mean(axis=2)
SE = FullyConnected("fc0({})".format(name),
SE,
output_dim=(k // group)**2 * group,
nonlinearity=ReLU())
SE = FullyConnected("fc1({})".format(name),
SE,
output_dim=(k // group)**2 * group,
nonlinearity=Sigmoid())
print(SE.name)
SE = SE.reshape(cur_lay.shape[0] * group, k // group, k // group, 1, 1)
preshape = cur_lay.shape
cur_lay = cur_lay.reshape(1, cur_lay.shape[0] * cur_lay.shape[1],
cur_lay.shape[2], cur_lay.shape[3])
cur_lay = Conv2D("conv({})".format(name),
cur_lay,
kernel_shape=1,
stride=1,
padding=0,
W=SE,
nonlinearity=Identity())
cur_lay = cur_lay.reshape(preshape)
#cur_lay = cur_lay * SE.dimshuffle(0, 1, 'x', 'x')
lay = Concat([lay, cur_lay], axis=1)
return lay
def dense_block(inp, k, l):
lay = inp
lis_w = []
for i in range(l):
cur_lay, w = bn_relu_conv(lay, 3, 1, 1, k, True, True)
lis_w.append(w)
lay = Concat([lay, cur_lay], axis=1)
return lay, lis_w
def dense_block(inp, k, l): lay = inp pre_lay = 0 for i in range(l): cur_lay = pre_lay + bn_relu_conv(lay, 3, 1, 1, k, True, True) pre_lay = bn_relu_conv(cur_lay, 3, 1, 1, k, True, True) lay = Concat([lay, cur_lay], axis = 1) return lay
def b_resize(name, inp, rate=0.8):
#inp = ConstProvider([[[[1, 2], [3, 4]]]], dtype = np.float32)
f_size = inp.partial_shape[2]
l = int(f_size * rate)
s = [[0, l], [f_size - l, f_size]]
ar0 = Linspace(0, inp.shape[0], inp.shape[0], endpoint=False)
ar0 = ar0.add_axis(1).add_axis(2).add_axis(3).broadcast(
inp.shape).add_axis(4)
ar1 = Linspace(0, inp.shape[1], inp.shape[1], endpoint=False)
ar1 = ar1.add_axis(0).add_axis(2).add_axis(3).broadcast(
inp.shape).add_axis(4)
fmaps = [inp]
for i in range(4):
xx = s[i % 2]
yy = s[i // 2]
#x = Linspace(xx[0], xx[1], f_size, endpoint = False)
#y = Linspace(yy[0], yy[1], f_size, endpoint = False)
x = ConstProvider(np.linspace(xx[0], xx[1], f_size, endpoint=False))
y = ConstProvider(np.linspace(yy[0], yy[1], f_size, endpoint=False))
fx, fy = Floor(x), Floor(y)
cx, cy = Ceil(x), Ceil(y)
nfmaps = []
for sx in range(2):
for sy in range(2):
ix = fx if sx == 0 else cx
iy = fy if sy == 0 else cy
bx = (cx - x + Equal(fx, cx) if sx == 0 else x - fx)
by = (cy - y + Equal(fy, cy) if sy == 0 else y - fy)
arx = ix.add_axis(0).add_axis(0).add_axis(3).broadcast(
inp.shape).add_axis(4)
ary = iy.add_axis(0).add_axis(0).add_axis(0).broadcast(
inp.shape).add_axis(4)
idxmap = Astype(Concat([ar0, ar1, arx, ary], axis=4), np.int32)
sample = IndexingRemap(inp, idxmap)
sample *= bx.dimshuffle('x', 'x', 0, 'x') * by.dimshuffle(
'x', 'x', 'x', 0)
nfmaps.append(sample)
fmap = nfmaps[0]
for i in range(1, 4):
fmap += nfmaps[i]
fmaps.append(fmap)
fmap = Concat(fmaps, axis=1)
return fmap
def dense_block(inp, k, l, j):
lay = inp
for i in range(l):
idx = j * l + i + 1
chl = lay.partial_shape[1]
p = idx & -idx
p = max(0, chl - p * k)
sh_lay = lay[:, p:, :, :]
cur_lay = bn_relu_conv(sh_lay, 3, 1, 1, k, True, True)
lay = Concat([lay, cur_lay], axis=1)
return lay
def dense_block(inp, k, l):
lay = inp
pre_lay = 0
for i in range(l):
chl = lay.partial_shape[1]
sh_lay = lay.reshape(lay.partial_shape[0], chl // k, k,
lay.partial_shape[2], lay.partial_shape[3])
sh_lay = sh_lay.dimshuffle(0, 2, 1, 3, 4)
sh_lay = sh_lay.reshape(lay.shape)
cur_lay = bn_relu_conv(sh_lay, 3, 1, 1, k, True, True, group=k // 4)
lay = Concat([lay, cur_lay], axis=1)
return lay
def dense_block(inp, k, l):
lay = inp
for i in range(l):
cur_lay = bn_relu_conv(lay, 3, 1, 1, k, True, True)
name = cur_lay.name
#G.P.
SE = cur_lay.mean(axis=3).mean(axis=2)
SE = FullyConnected("fc0({})".format(name),
SE,
output_dim=k,
nonlinearity=ReLU())
SE = FullyConnected("fc1({})".format(name),
SE,
output_dim=k,
nonlinearity=Sigmoid())
cur_lay = cur_lay * SE.dimshuffle(0, 1, 'x', 'x')
lay = Concat([lay, cur_lay], axis=1)
return lay
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]):
inp = Conv2D(
name + "conv", inp, kernel_shape = 3, stride = 1, padding = 1,
output_nr_channel = ker_shape**2,
W = G(mean = 0, std = ((1) / (ker_shape**2 * inp.partial_shape[1]))**0.5),
nonlinearity = Identity()
)
inp = BN(name + "BN", inp, eps = 1e-9)
global idx
#idx += 1
gamma = 0.001
offsetx = inp.partial_shape[2] * Conv2D(
"conv{}_offsetx".format(idx + 1), inp, kernel_shape = ker_shape, stride = stride,
padding = padding,
output_nr_channel = ker_shape**2,
W = G(mean = 0, std = gamma / (ker_shape**2 * inp.partial_shape[2])),
nonlinearity = Identity()
)
offsety = inp.partial_shape[3] * Conv2D(
"conv{}_offsety".format(idx + 1), inp, kernel_shape = ker_shape, stride = stride,
padding = padding,
output_nr_channel = ker_shape**2,
W = G(mean = 0, std = gamma / (ker_shape**2 * inp.partial_shape[3])),
nonlinearity = Identity()
)
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 dfpooling(name, inp, window = 2, padding = 0, dx = [0, 1], dy = [0, 1]):
#inp = ConstProvider([[[[1, 2], [3, 4]]]], dtype = np.float32)
"""
Add a new conv&bn to insure that the scale of the feature map is variance 1.
"""
ker_shape = window
stride = window
offsetlay = Conv2D(
name + "conv", 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 + "BN", offsetlay, eps = 1e-9)
offsetx = Conv2D(
name + "conv1x", 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 = Conv2D(
name + "conv1y", 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()
)
offset = Concat([offsetx, offsety], axis = 1)
ndim = ker_shape**2 * offsetx.partial_shape[2] * offsetx.partial_shape[3] * 2
offset = FullyConnected(
name + "offset", offsetx, output_dim = ndim,
W = G(mean = 0, std = (1 / ndim)**2),
#W = C(0),
b = C(0),
nonlinearity = Identity()
)
offsetx = offset[:, :ndim // 2].reshape(offsetx.shape)
offsety = offset[:, ndim // 2:].reshape(offsety.shape)
"""
offsetx = FullyConnected(
name + "offsetx", offsetx, output_dim = ndim,
W = G(mean = 0, std = gamma / ndim),
b = C(0),
nonlinearity = Identity()
)
offsetx = offsetx.reshape(offsety.shape)
offsety = FullyConnected(
name + "offsety", offsety, output_dim = ndim,
W = G(mean = 0, std = gamma / ndim),
b = C(0),
nonlinearity = Identity()
)
offsety = offsety.reshape(offsetx.shape)
print(offsety.partial_shape)
"""
#offsetx = ZeroGrad(offsetx)
#offsety = ZeroGrad(offsety)
outputs = []
for sx in range(2):
for sy in range(2):
if sx == 0:
ofx = Floor(offsetx)
bilx = 1 - (offsetx - ofx)
else:
ofx = Ceil(offsetx)
bilx = 1 - (ofx - offsetx)
if sy == 0:
ofy = Floor(offsety)
bily = 1 - (offsety - ofy)
else:
ofy = Ceil(offsety)
bily = 1 - (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.shape[2], ofx.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.partial_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.partial_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 Pooling2D(name, output, window = 2, mode = "AVERAGE")
def dense_block(inp, k, l, B):
lay = inp
for i in range(l):
cur_lay = dense_layer(lay, k, B)
lay = Concat([lay, cur_lay], axis=1)
return lay
def dfpooling(name, inp, window=2, padding=0, dx=[0, 1], dy=[0, 1]):
#inp = ConstProvider([[[[1, 2], [3, 4]]]], dtype = np.float32)
ker_shape = window
stride = window
gamma = 0.1
offsetx = gamma * inp.partial_shape[2] * Conv2D(name + "offsetx",
inp,
kernel_shape=ker_shape,
stride=stride,
padding=padding,
output_nr_channel=ker_shape
**2,
W=C(0),
nonlinearity=Identity())
offsety = gamma * inp.partial_shape[3] * Conv2D(name + "offsety",
inp,
kernel_shape=ker_shape,
stride=stride,
padding=padding,
output_nr_channel=ker_shape
**2,
W=C(0),
nonlinearity=Identity())
outputs = []
for sx in range(2):
for sy in range(2):
if sx == 0:
ofx = Floor(offsetx)
bilx = offsetx - ofx + Equal(Floor(offsetx), Ceil(offsetx))
else:
ofx = Ceil(offsetx)
bilx = ofx - offsetx
if sy == 0:
ofy = Floor(offsety)
bily = offsety - ofy + Equal(Floor(offsety), Ceil(offsety))
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.partial_shape[0], ker_shape, ker_shape,
inp.partial_shape[2], inp.partial_shape[3])
inp = inp.dimshuffle(0, 3, 1, 4, 2)
inp = inp.reshape(inp.partial_shape[0],
inp.partial_shape[1] * inp.partial_shape[2],
inp.partial_shape[3] * inp.partial_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.partial_shape[2] + ofy
arg_fea = arg_fea.reshape(arg_fea.partial_shape[0],
arg_fea.partial_shape[1], -1)
of = of.reshape(ofx.partial_shape[0], -1)
of = of.dimshuffle(0, 'x', 1)
#of = Concat([of] * arg_fea.partial_shape[1], axis = 1)
of = of.broadcast((of.partial_shape[0], arg_fea.partial_shape[1],
of.partial_shape[2]))
arx = Linspace(0,
arg_fea.partial_shape[0],
arg_fea.partial_shape[0],
endpoint=False)
arx = arx.add_axis(1).add_axis(2).broadcast(of.shape)
ary = Linspace(0,
arg_fea.partial_shape[1],
arg_fea.partial_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.partial_shape[0],
inp.partial_shape[1],
bilx.partial_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 Pooling2D(name, output, window=2, mode="AVERAGE")
