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

[Paddle-TRT] fix_ernie #44502

Closed
wants to merge 4 commits into from
Closed

[Paddle-TRT] fix_ernie #44502

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
2162d77
fix_ernie_all
zhoutianzi666 Jul 21, 2022
689d0e5
fix_ernie_all
zhoutianzi666 Jul 21, 2022
7014b69
fix_ernie_all
zhoutianzi666 Jul 21, 2022
f1887ee
fix_ernie_all
zhoutianzi666 Jul 21, 2022
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
32 changes: 22 additions & 10 deletions paddle/fluid/inference/tensorrt/convert/skip_layernorm.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,12 +26,29 @@ class SkipLayerNormOpConverter : public OpConverter {
void operator()(const framework::proto::OpDesc& op,
const framework::Scope& scope,
bool test_mode) override {
#if IS_TRT_VERSION_GE(6000)
VLOG(4) << "convert fused skip layernorm op to tensorrt layer";
framework::OpDesc op_desc(op, nullptr);
// Declare inputs
auto* input1 = engine_->GetITensor(op_desc.Input("X")[0]);
auto* input2 = engine_->GetITensor(op_desc.Input("Y")[0]);

// We should expand input1 to input2's nbDims
if (input1->getDimensions().nbDims < input2->getDimensions().nbDims) {
nvinfer1::Dims reshape_input1_dim;
reshape_input1_dim.nbDims = input2->getDimensions().nbDims;
for (int i = 0; i < reshape_input1_dim.nbDims; i++) {
if (i < input1->getDimensions().nbDims) {
reshape_input1_dim.d[i] = 0;
} else {
reshape_input1_dim.d[i] = 1;
}
}
auto* reshape_input1_layer =
TRT_ENGINE_ADD_LAYER(engine_, Shuffle, *input1);
reshape_input1_layer->setReshapeDimensions(reshape_input1_dim);
input1 = reshape_input1_layer->getOutput(0);
}

std::vector<nvinfer1::ITensor*> inputs;
inputs.push_back(input1);
inputs.push_back(input2);
Expand Down Expand Up @@ -74,10 +91,10 @@ class SkipLayerNormOpConverter : public OpConverter {
bias_weight.values,
GetPluginFieldType(bias_weight.type),
static_cast<int32_t>(bias_weight.count)},
{ "gamma",
scale_weight.values,
GetPluginFieldType(scale_weight.type),
static_cast<int32_t>(scale_weight.count) }};
{"gamma",
scale_weight.values,
GetPluginFieldType(scale_weight.type),
static_cast<int32_t>(scale_weight.count)}};
nvinfer1::PluginFieldCollection* pluginPtr =
static_cast<nvinfer1::PluginFieldCollection*>(
malloc(sizeof(*pluginPtr) +
Expand Down Expand Up @@ -178,11 +195,6 @@ class SkipLayerNormOpConverter : public OpConverter {

auto output_name = op_desc.Output("Out")[0];
RreplenishLayerAndOutput(layer, "skip_layernorm", {output_name}, test_mode);
#else
PADDLE_THROW(platform::errors::Fatal(
"You are running the TRT Dynamic Shape mode, need to confirm that "
"your TRT version is no less than 6.0"));
#endif
}
};

Expand Down