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

[Paddle-Inference] Add IPluginV2Layer : AddPluginV2IOExt #36493

Merged
merged 1 commit into from
Oct 18, 2021
Merged

[Paddle-Inference] Add IPluginV2Layer : AddPluginV2IOExt #36493

Show file tree
Hide file tree
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
13 changes: 7 additions & 6 deletions paddle/fluid/inference/tensorrt/engine.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -135,12 +135,6 @@ void TensorRTEngine::FreezeNetwork() {
}
for (int j = 0; j < layer->getNbOutputs(); j++) {
auto *temp_out = layer->getOutput(j);
if (temp_out->isNetworkOutput()) {
VLOG(1) << "Layer(Name: " << layer->getName()
<< ") is set to float32 because its output("
<< temp_out->getName() << ") is the output of the network.";
return false;
}
if (!temp_out->dynamicRangeIsSet()) {
VLOG(1) << "Layer(Name: " << layer->getName()
<< ") is set to float32 because its output("
Expand Down Expand Up @@ -357,6 +351,13 @@ nvinfer1::IPluginV2Layer *TensorRTEngine::AddPluginV2Ext(
return network()->addPluginV2(inputs, num_inputs, *plugin);
}

nvinfer1::IPluginV2Layer *TensorRTEngine::AddPluginV2IOExt(
nvinfer1::ITensor *const *inputs, int num_inputs,
nvinfer1::IPluginV2IOExt *plugin) {
owned_plugin_v2ioext_.emplace_back(plugin);
return network()->addPluginV2(inputs, num_inputs, *plugin);
}

void TensorRTEngine::freshDeviceId() {
int count;
cudaGetDeviceCount(&count);
Expand Down
6 changes: 6 additions & 0 deletions paddle/fluid/inference/tensorrt/engine.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -323,6 +323,10 @@ class TensorRTEngine {
int num_inputs,
plugin::PluginTensorRTV2Ext* plugin);

nvinfer1::IPluginV2Layer* AddPluginV2IOExt(nvinfer1::ITensor* const* inputs,
int num_inputs,
nvinfer1::IPluginV2IOExt* plugin);

void SetTensorDynamicRange(nvinfer1::ITensor* tensor, float range) {
quant_dynamic_range_[tensor] = range;
}
Expand Down Expand Up @@ -429,6 +433,7 @@ class TensorRTEngine {
bool with_ernie() { return with_ernie_; }
bool disable_trt_plugin_fp16() { return disable_trt_plugin_fp16_; }
bool with_dynamic_shape() { return with_dynamic_shape_; }
AnalysisConfig::Precision precision() { return precision_; }

#if IS_TRT_VERSION_GE(6000)
nvinfer1::IPluginV2Layer* AddDynamicPlugin(
Expand Down Expand Up @@ -550,6 +555,7 @@ class TensorRTEngine {

std::vector<std::unique_ptr<plugin::PluginTensorRT>> owned_plugin_;
std::vector<std::unique_ptr<plugin::PluginTensorRTV2Ext>> owned_plugin_v2ext_;
std::vector<std::unique_ptr<nvinfer1::IPluginV2IOExt>> owned_plugin_v2ioext_;

// TensorRT related internal members
template <typename T>
Expand Down