Inference on a GPU

GPU Inference Application

Use C++ Interface to Load a MindIR File for Inferencing

Inference Directory Structure

Create a directory to store the inference code project, for example, /home/mindspore_sample/gpu_resnet50_inference_sample. You can download the sample code from the official website. The model directory is used to store the exported MindIR model file. The directory structure of the inference code project is as follows:

└─gpu_resnet50_inference_sample
    ├── build.sh                          // Build script
    ├── CMakeLists.txt                    // CMake script
    ├── README.md                         // Usage description
    ├── src
    │   └── main.cc                       // Main function
    └── model
        └── resnet50_imagenet.mindir      // MindIR model file

Inference Code

Namespaces that reference mindspore.

using mindspore::Context;
using mindspore::Serialization;
using mindspore::Model;
using mindspore::Status;
using mindspore::ModelType;
using mindspore::GraphCell;
using mindspore::kSuccess;
using mindspore::MSTensor;

Initialize the environment, specify the hardware platform used for inference, and set DeviceID.

Set the hardware to GPU, set DeviceID to 0 and Precision Mode to “fp16”. The code example is as follows:

auto gpu_device_info = std::make_shared<mindspore::GPUDeviceInfo>();
gpu_device_info->SetDeviceID(device_id);
gpu_device_info->SetPrecisionMode("fp16");
context->MutableDeviceInfo().push_back(gpu_device_info);

Load the model file.

// Load the MindIR model.
mindspore::Graph graph;
Serialization::Load(mindir_path, ModelType::kMindIR, &graph);
// Build a model using a graph.
ms::Model model;
model.Build(ms::GraphCell(graph), context);

Obtain the input information required by the model.

std::vector<ms::MSTensor> model_inputs = model->GetInputs();

Start inference.

// Create an output vector.
std::vector<ms::MSTensor> outputs;
// Create an input vector.
std::vector<ms::MSTensor> inputs;
inputs.emplace_back(model_inputs[0].Name(), model_inputs[0].DataType(), model_inputs[0].Shape(),
                    image.Data().get(), image.DataSize());
// Call the Predict function of the model for inference.
ret = model.Predict(inputs, &outputs);

Introduce to Building Script

Add the header file search path for the compiler:

option(MINDSPORE_PATH "mindspore install path" "")
include_directories(${MINDSPORE_PATH})
include_directories(${MINDSPORE_PATH}/include)

Search for the required dynamic library in MindSpore.

find_library(MS_LIB libmindspore.so ${MINDSPORE_PATH}/lib)

Use the specified source file to generate the target executable file and link the target file to the MindSpore library.

add_executable(main src/main.cc)
target_link_libraries(main ${MS_LIB})

For details, see https://gitee.com/mindspore/docs/blob/r1.5/docs/sample_code/gpu_resnet50_inference_sample/CMakeLists.txt

Building Inference Code

Go to the project directory gpu_resnet50_inference_sample and modify the pip3 in the build.sh based on the actual situation. And then execute the building script.

bash build.sh

After building, the executable main file is generated in gpu_resnet50_inference_sample/out.

Performing Inference and Viewing the Result

After completing the preceding operations, you can learn how to perform inference.

Log in to the GPU environment, and create the model directory to store the resnet50_imagenet.mindir file, for example, /home/mindspore_sample/gpu_resnet50_inference_sample/model.

Set the environment variable base on the actual situation, where the TensorRT is an optional configuration item. It is recommended to add TensorRT path to LD_LIBRARY_PATH to improve mode inference performance.

export LD_PRELOAD=/home/miniconda3/lib/libpython37m.so
export LD_LIBRARY_PATH=/usr/local/TensorRT-7.2.2.3/lib/:$LD_LIBRARY_PATH

Then, perform inference for 1000 times after 10 times warmup.

cd out/
./main ../model/resnet50_imagenet.mindir 1000 10

In this example, we print the inference delay for per step and average step.

Start to load model..
Load model successuflly
Start to warmup..
Warmup finished
Start to infer..
step 0 cost 1.54004ms
step 1 cost 1.5271ms
... ...
step 998 cost 1.30688ms
step 999 cost 1.30493ms
infer finished.
=================Average inference time: 1.35195 ms

Notices

During the training process, some networks set operator precision to FP16 artificially. For example, the Bert mode set the Dense and LayerNorm to FP16:

class BertOutput(nn.Cell):
    def __init__(self,
                 in_channels,
                 out_channels,
                 initializer_range=0.02,
                 dropout_prob=0.1,
                 compute_type=mstype.float32):
        super(BertOutput, self).__init__()
        # Set the nn.Dense to fp16.
        self.dense = nn.Dense(in_channels, out_channels,
                              weight_init=TruncatedNormal(initializer_range)).to_float(compute_type)
        self.dropout = nn.Dropout(1 - dropout_prob)
        self.dropout_prob = dropout_prob
        self.add = P.Add()
        # Set the nn.LayerNorm to fp16.
        self.layernorm = nn.LayerNorm((out_channels,)).to_float(compute_type)
        self.cast = P.Cast()
        ... ...

It is recommended that export the MindIR model with fp32 precision mode before deploying inference. If you want to further improve the inference performance, you can set precision_mode is to “fp16”.

Some inference scripts may introduce some unique network structures in the training process. For example, the model requires the image label, which are transmitted to the network output directly. It is suggested to delete this part of operators and then export MindIR model to improve inference performance.

Inference Using an ONNX File

Generate a model in ONNX format on the training platform. For details, see Export ONNX Model.
Perform inference on a GPU by referring to the runtime or SDK document. For example, use TensorRT to perform inference on the NVIDIA GPU. For details, see TensorRT backend for ONNX.