Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix:transform the data from cpu to gpu when trt is used #37427

Merged
merged 1 commit into from
Nov 24, 2021
Merged

fix:transform the data from cpu to gpu when trt is used #37427

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions paddle/fluid/operators/tensorrt/tensorrt_engine_op.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@
#include <utility>
#include <vector>

#include "paddle/fluid/framework/data_device_transform.h"
#include "paddle/fluid/framework/executor.h"
#include "paddle/fluid/framework/op_registry.h"
#include "paddle/fluid/framework/operator.h"
Expand Down Expand Up @@ -421,6 +422,13 @@ class TensorRTEngineOp : public framework::OperatorBase {
// convert input and copy to TRT engine's buffer
auto &t =
inference::analysis::GetFromScope<framework::LoDTensor>(scope, x);
// check the input_tensor
if (!platform::is_gpu_place(t.place())) {
framework::Tensor out;
platform::CUDAPlace dst_place;
framework::TransDataDevice(t, dst_place, &out);
t.ShareDataWith(out);
}
auto t_shape = framework::vectorize<int64_t>(t.dims());
const int bind_index = engine->engine()->getBindingIndex(x.c_str());
PADDLE_ENFORCE_LT(
Expand Down