wehub-resource-sync
200 lines
6.5 KiB
Python
200 lines
6.5 KiB
Python
# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# 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.
|
|
|
|
import collections
|
|
import unittest
|
|
from functools import reduce
|
|
|
|
import numpy as np
|
|
|
|
import paddle
|
|
from paddle import base
|
|
from paddle.nn.utils import remove_weight_norm, weight_norm
|
|
|
|
|
|
class TestDygraphWeightNorm(unittest.TestCase):
|
|
def setUp(self):
|
|
self.init_test_case()
|
|
self.set_data()
|
|
|
|
def init_test_case(self):
|
|
self.batch_size = 3
|
|
self.data_desc = (['x', [2, 3, 3]],)
|
|
self.dim = None
|
|
|
|
def set_data(self):
|
|
self.data = collections.OrderedDict()
|
|
for desc in self.data_desc:
|
|
data_name = desc[0]
|
|
data_shape = desc[1]
|
|
data_value = np.random.random(
|
|
size=[self.batch_size, *data_shape]
|
|
).astype('float32')
|
|
self.data[data_name] = data_value
|
|
|
|
def norm_except_dim(self, w, dim=None):
|
|
shape = w.shape
|
|
ndims = len(shape)
|
|
shape_numel = reduce(lambda x, y: x * y, shape, 1)
|
|
if dim == -1:
|
|
return np.linalg.norm(w, axis=None, keepdims=True).flatten()
|
|
elif dim == 0:
|
|
tile_shape = list(w.shape)
|
|
tile_shape[0] = 1
|
|
w_matrix = np.reshape(w, (shape[0], shape_numel // shape[0]))
|
|
return np.linalg.norm(w_matrix, axis=1, keepdims=True)
|
|
elif dim == (ndims - 1):
|
|
w_matrix = np.reshape(w, (shape_numel // shape[-1], shape[-1]))
|
|
return np.linalg.norm(w_matrix, axis=0, keepdims=True)
|
|
else:
|
|
perm = list(range(ndims))
|
|
perm_ori = list(range(ndims))
|
|
perm[0] = dim
|
|
perm[dim] = 0
|
|
p_transposed = np.transpose(w, perm)
|
|
return self.norm_except_dim(p_transposed, 0)
|
|
|
|
def weight_normalize(self, w, dim=None):
|
|
shape = w.shape
|
|
ndims = len(shape)
|
|
shape_numel = reduce(lambda x, y: x * y, shape, 1)
|
|
v = w
|
|
g = self.norm_except_dim(w, dim)
|
|
g_mul = g
|
|
|
|
if dim == -1:
|
|
v_norm = v / (np.linalg.norm(v, axis=None, keepdims=True))
|
|
elif dim == 0:
|
|
w_matrix = np.reshape(w, (shape[0], shape_numel // shape[0]))
|
|
v_norm = v / np.linalg.norm(w_matrix, axis=1)
|
|
v_norm = np.reshape(v_norm, shape)
|
|
g = np.squeeze(g, axis=1)
|
|
elif dim == (ndims - 1):
|
|
w_matrix = np.reshape(w, (shape_numel // shape[-1], shape[-1]))
|
|
v_norm = v / np.linalg.norm(w_matrix, axis=0, keepdims=True)
|
|
v_norm = np.reshape(v_norm, shape)
|
|
else:
|
|
perm = list(range(ndims))
|
|
perm[0] = dim
|
|
perm[dim] = 0
|
|
p_transposed = np.transpose(v, perm)
|
|
transposed_shape = p_transposed.shape
|
|
transposed_shape_numel = reduce(
|
|
lambda x, y: x * y, transposed_shape
|
|
)
|
|
p_matrix = np.reshape(
|
|
p_transposed,
|
|
(
|
|
p_transposed.shape[0],
|
|
transposed_shape_numel // p_transposed.shape[0],
|
|
),
|
|
)
|
|
v_norm = v / np.expand_dims(
|
|
np.expand_dims(
|
|
np.linalg.norm(p_matrix, axis=1, keepdims=True), axis=0
|
|
),
|
|
axis=(ndims - 1),
|
|
)
|
|
v_norm = np.reshape(v_norm, transposed_shape)
|
|
v_norm = np.transpose(v_norm, perm)
|
|
g = np.squeeze(g, axis=1)
|
|
if dim == 1:
|
|
eaxis = 2
|
|
elif dim == 2:
|
|
eaxis = 1
|
|
g_mul = np.expand_dims(
|
|
np.expand_dims(np.expand_dims(g, axis=0), axis=eaxis),
|
|
axis=(ndims - 1),
|
|
)
|
|
w = g_mul * v_norm
|
|
return g, v
|
|
|
|
def test_check_output(self):
|
|
base.enable_imperative()
|
|
linear = paddle.nn.Conv2D(2, 3, 3)
|
|
before_weight = linear.weight.numpy()
|
|
if self.dim is None:
|
|
self.dim = -1
|
|
|
|
if self.dim != -1:
|
|
self.dim = (self.dim + len(before_weight)) % len(before_weight)
|
|
wn = weight_norm(linear, dim=self.dim)
|
|
outputs = []
|
|
for name, data in self.data.items():
|
|
output = linear(paddle.to_tensor(data))
|
|
outputs.append(output.numpy())
|
|
after_weight = linear.weight
|
|
self.actual_outputs = [linear.weight_g.numpy(), linear.weight_v.numpy()]
|
|
|
|
expect_output = self.weight_normalize(before_weight, self.dim)
|
|
|
|
for expect, actual in zip(expect_output, self.actual_outputs):
|
|
np.testing.assert_allclose(
|
|
np.array(actual), expect, rtol=1e-05, atol=0.001
|
|
)
|
|
|
|
|
|
class TestDygraphWeightNormCase1(TestDygraphWeightNorm):
|
|
def init_test_case(self):
|
|
self.batch_size = 3
|
|
self.data_desc = (['x', [2, 3, 3]],)
|
|
self.dim = 0
|
|
|
|
|
|
class TestDygraphWeightNormCase2(TestDygraphWeightNorm):
|
|
def init_test_case(self):
|
|
self.batch_size = 3
|
|
self.data_desc = (['x', [2, 3, 3]],)
|
|
self.dim = 1
|
|
|
|
|
|
class TestDygraphWeightNormCase3(TestDygraphWeightNorm):
|
|
def init_test_case(self):
|
|
self.batch_size = 3
|
|
self.data_desc = (['x', [2, 3, 3]],)
|
|
self.dim = 3
|
|
|
|
|
|
class TestDygraphWeightNormCase4(TestDygraphWeightNorm):
|
|
def init_test_case(self):
|
|
self.batch_size = 3
|
|
self.data_desc = (['x', [2, 3, 3]],)
|
|
self.dim = -3
|
|
|
|
|
|
class TestDygraphRemoveWeightNorm(unittest.TestCase):
|
|
def setUp(self):
|
|
self.init_test_case()
|
|
|
|
def init_test_case(self):
|
|
self.batch_size = 3
|
|
self.data_desc = (['x', [2, 3, 3]],)
|
|
self.dim = None
|
|
|
|
def test_check_output(self):
|
|
base.enable_imperative()
|
|
linear = paddle.nn.Conv2D(2, 3, 3)
|
|
before_weight = linear.weight
|
|
wn = weight_norm(linear, dim=self.dim)
|
|
rwn = remove_weight_norm(linear)
|
|
after_weight = linear.weight
|
|
np.testing.assert_allclose(
|
|
before_weight.numpy(), after_weight.numpy(), rtol=1e-05, atol=0.001
|
|
)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
paddle.enable_static()
|
|
unittest.main()
|