未验证 提交 4a0a8701 编写于 作者: H Huihuang Zheng 提交者: GitHub

[dy2stat] Set shape for linspace to Fix dy2stat for GridGenerator Model (#29173)

GridGenerator model failed because the output shape of `linspace` is (-1). The reason is that C++ InferShape fixes the shape to (-1): 

https://github.com/PaddlePaddle/Paddle/blob/5da3d514ebaa6fffd48c4a2e6bb5b16268dae92e/paddle/fluid/operators/linspace_op.cc#L49

We cannot set the shape in C++ infer shape because this Tensor may not be initialized during compile time, but when input `num` of `linspace` is an integer, we know the shape at compiler time. This PR simply set the shape in Python and add GridGenerator as unittest.
 
上级 cb680c80
......@@ -1477,6 +1477,8 @@ def linspace(start, stop, num, dtype=None, name=None):
'Num': tensor_num},
attrs={'dtype': dtype},
outputs={'Out': [out]})
if isinstance(num, int):
out.desc.set_shape((num, ))
return out
......
# copyright (c) 2020 PaddlePaddle Authors. All Rights Reserve.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import paddle
from paddle import nn, ParamAttr
import numpy as np
import unittest
np.random.seed(2020)
paddle.seed(2020)
class GridGenerator(nn.Layer):
def __init__(self, in_channels, num_fiducial):
super(GridGenerator, self).__init__()
self.eps = 1e-6
self.F = num_fiducial
initializer = nn.initializer.Constant(value=0.0)
param_attr = ParamAttr(learning_rate=0.0, initializer=initializer)
bias_attr = ParamAttr(learning_rate=0.0, initializer=initializer)
self.fc = nn.Linear(
in_channels, 6, weight_attr=param_attr, bias_attr=bias_attr)
@paddle.jit.to_static(input_spec=[
paddle.static.InputSpec(
shape=[None, 3, 32, 100], dtype='float32'), paddle.static.InputSpec(
shape=[32, 100], dtype='float32')
])
def forward(self, batch_C_prime, I_r_size):
"""
Generate the grid for the grid_sampler.
Args:
batch_C_prime: the matrix of the geometric transformation
I_r_size: the shape of the input image
Return:
batch_P_prime: the grid for the grid_sampler
"""
C = self.build_C_paddle()
return C
def build_C_paddle(self):
""" Return coordinates of fiducial points in I_r; C """
F = self.F
ctrl_pts_x = paddle.linspace(-1.0, 1.0, int(F / 2))
ctrl_pts_y_top = -1 * paddle.ones([int(F / 2)])
ctrl_pts_y_bottom = paddle.ones([int(F / 2)])
ctrl_pts_top = paddle.stack([ctrl_pts_x, ctrl_pts_y_top], axis=1)
ctrl_pts_bottom = paddle.stack([ctrl_pts_x, ctrl_pts_y_bottom], axis=1)
C = paddle.concat([ctrl_pts_top, ctrl_pts_bottom], axis=0)
return C
def build_P_paddle(self, I_r_size):
I_r_width, I_r_height = I_r_size
I_r_grid_x = paddle.divide(
(paddle.arange(-I_r_width, I_r_width, 2).astype('float32') + 1.0),
paddle.to_tensor(I_r_width).astype('float32'))
I_r_grid_y = paddle.divide(
(paddle.arange(-I_r_height, I_r_height, 2).astype('float32') + 1.0),
paddle.to_tensor(I_r_height).astype('float32'))
P = paddle.stack(paddle.meshgrid(I_r_grid_x, I_r_grid_y), axis=2)
P = paddle.transpose(P, perm=[1, 0, 2])
return P.reshape([-1, 2])
def build_inv_delta_C_paddle(self, C):
""" Return inv_delta_C which is needed to calculate T """
F = self.F
hat_C = paddle.zeros((F, F), dtype='float32')
for i in range(0, F):
for j in range(i, F):
if i == j:
hat_C[i, j] = 1
else:
r = paddle.norm(C[i] - C[j])
hat_C[i, j] = r
hat_C[j, i] = r
hat_C = (hat_C**2) * paddle.log(hat_C)
delta_C = paddle.concat(
[
paddle.concat(
[paddle.ones((F, 1)), C, hat_C], axis=1),
paddle.concat(
[paddle.zeros((2, 3)), paddle.transpose(
C, perm=[1, 0])],
axis=1), paddle.concat(
[paddle.zeros((1, 3)), paddle.ones((1, F))], axis=1)
],
axis=0)
inv_delta_C = paddle.inverse(delta_C)
return inv_delta_C
def build_P_hat_paddle(self, C, P):
F = self.F
eps = self.eps
n = P.shape[0]
P_tile = paddle.tile(paddle.unsqueeze(P, axis=1), (1, F, 1))
C_tile = paddle.unsqueeze(C, axis=0)
P_diff = P_tile - C_tile
rbf_norm = paddle.norm(P_diff, p=2, axis=2, keepdim=False)
rbf = paddle.multiply(
paddle.square(rbf_norm), paddle.log(rbf_norm + eps))
P_hat = paddle.concat([paddle.ones((n, 1)), P, rbf], axis=1)
return P_hat
def get_expand_tensor(self, batch_C_prime):
B, H, C = batch_C_prime.shape
batch_C_prime = batch_C_prime.reshape([B, H * C])
batch_C_ex_part_tensor = self.fc(batch_C_prime)
batch_C_ex_part_tensor = batch_C_ex_part_tensor.reshape([-1, 3, 2])
return batch_C_ex_part_tensor
class TestGridGenerator(unittest.TestCase):
def setUp(self):
self.x = paddle.uniform(shape=[1, 20, 2], dtype='float32')
def _run(self, to_static):
prog_trans = paddle.jit.ProgramTranslator()
prog_trans.enable(to_static)
net = GridGenerator(40, 20)
ret = net(self.x, [32, 100])
return ret.numpy()
def test_to_static(self):
st_out = self._run(to_static=True)
dy_out = self._run(to_static=False)
np.testing.assert_allclose(st_out, dy_out)
if __name__ == '__main__':
unittest.main()
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册