未验证 提交 e58ac121 编写于 作者: Z Zhanlue Yang 提交者: GitHub

Bug fix for snapshotting VariableWrapper with initialized tensor but e… (#37410)

* Bug fix for snapshoting VariableWrapper with initialized tensor but empty allocation

* Added unittest for inplace&clear_gradient
上级 90dad8b2
......@@ -347,16 +347,28 @@ class TracedGradOp {
// Use original var_wrapper if its inplace_version is not
// changed. Otherwise, it will affect the accuracy of the model
// results and affect double grad.
if (!var_wrapper->MutableVar()->IsInitialized() ||
var_wrapper->InplaceVersionSnapshot() ==
if (!var_wrapper->MutableVar()->IsInitialized()) {
return var_wrapper;
} else if (var_wrapper->InplaceVersionSnapshot() ==
var_wrapper->MutableVar()->CurrentInplaceVersion()) {
return var_wrapper;
} else {
} else if (var_wrapper->MutableVar()->IsType<framework::LoDTensor>() ||
var_wrapper->MutableVar()->IsType<framework::SelectedRows>()) {
auto* tensor =
var_wrapper->MutableVar()->IsType<framework::LoDTensor>()
? var_wrapper->MutableVar()->GetMutable<framework::LoDTensor>()
: var_wrapper->MutableVar()
->GetMutable<framework::SelectedRows>()
->mutable_value();
if (!tensor->IsInitialized()) {
return var_wrapper;
}
}
VariableWrapper new_var_wrapper = *var_wrapper.get();
new_var_wrapper.ResetInplaceVersion();
return std::make_shared<VariableWrapper>(new_var_wrapper);
}
}
private:
const std::shared_ptr<GradOpNode>& node_;
......
# Copyright (c) 2021 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 numpy as np
import paddle
import paddle.fluid as fluid
from paddle import _C_ops
import unittest
paddle.disable_static()
def clear_grad(w, a):
@paddle.no_grad()
def warp(*_):
assert w.grad is not None
_C_ops.scale_(w.grad, 'scale', 0.5)
w.clear_gradient(False)
return warp
class TestInplaceAndClearGradient(unittest.TestCase):
def test(self):
paddle.set_device('cpu')
input_data = np.ones([2, 2]).astype('float32')
w = paddle.to_tensor(input_data, 'float32', stop_gradient=False)
_clear_grad = clear_grad(w, a="1")
w._register_backward_hook(_clear_grad)
for i in range(10):
out = _C_ops.scale(w, 'scale', 0.1)
out.backward()
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