Merge Master, update optimizer (#151)

* [Feature] add yolox ncnn (#29) * add yolox ncnn * add ncnn android performance of yolox * add ut * fix lint * fix None bugs for ncnn * test codecov * test codecov * add device * fix yapf * remove if-else for img shape * use channelshuffle optimize * change benchmark after channelshuffle * fix yapf * fix yapf * fuse continuous reshape * fix static shape deploy * fix code * drop pad * only static shape * fix static * fix docstring * Added mask overlay to output image, changed fprintf info messages to … (#55) * Added mask overlay to output image, changed fprintf info messages to stdout * Improved box filtering (filter area/score), make sure roi coordinates stay within bounds * clang-format * Support UNet in mmseg (#77) * Repeatdataset in train has no CLASSES & PALETTE * update result for unet * update docstring for mmdet * remove ppl for unet in docs * fix ort wrap about input type (#81) * Fix memleak (#86) * delete [] * fix build error when enble MMDEPLOY_ACTIVE_LEVEL * fix lint * [Doc] Nano benchmark and tutorial (#71) * add cls benchmark * add nano zh-cn benchmark and en tutorial * add device row * add doc path to index.rst * fix typo * [Fix] fix missing deploy_core (#80) * fix missing deploy_core * mv flag to demo * target link * [Docs] Fix links in Chinese doc (#84) * Fix docs in Chinese link * Fix links * Delete symbolic link and add links to html * delete files * Fix link * [Feature] Add docker files (#67) * add gpu and cpu dockerfile * fix lint * fix cpu docker and remove redundant * use pip instead * add build arg and readme * fix grammar * update readme * add chinese doc for dockerfile and add docker build to build.md * grammar * refine dockerfiles * add FAQs * update Dpplcv_DIR for SDK building * remove mmcls * add sdk demos * fix typo and lint * update FAQs * [Fix]fix check_env (#101) * fix check_env * update * Replace convert_syncbatchnorm in mmseg (#93) * replace convert_syncbatchnorm with revert_sync_batchnorm from mmcv * change logger * [Doc] Update FAQ for TensorRT (#96) * update FAQ * comment * [Docs]: Update doc for openvino installation (#102) * fix docs * fix docs * fix docs * fix mmcv version * fix docs * rm blank line * simplify non batch nms (#99) * [Enhacement] Allow test.py to save evaluation results (#108) * Add log file * Delete debug code * Rename logger * resolve comments * [Enhancement] Support mmocr v0.4+ (#115) * support mmocr v0.4+ * 0.4.0 -> 0.4.1 * fix onnxruntime wrapper for gpu inference (#123) * fix ncnn wrapper for ort-gpu * resolve comment * fix lint * Fix typo (#132) * lock mmcls version (#131) * [Enhancement] upgrade isort in pre-commit config (#141) * [Enhancement] upgrade isort in pre-commit config by refering to mmflow pr #87 * fix lint * remove .isort.cfg and put its known_third_party to setup.cfg * Fix ci for mmocr (#144) * fix mmocr unittests * remove useless * lock mmdet maximum version to 2.20 * pip install -U numpy * Fix capture_output (#125) Co-authored-by: hanrui1sensetime <[email protected]> Co-authored-by: Johannes L <[email protected]> Co-authored-by: RunningLeon <[email protected]> Co-authored-by: VVsssssk <[email protected]> Co-authored-by: lvhan028 <[email protected]> Co-authored-by: AllentDan <[email protected]> Co-authored-by: Yifan Zhou <[email protected]> Co-authored-by: 杨培文 (Yang Peiwen) <[email protected]> Co-authored-by: Semyon Bevzyuk <[email protected]>
open-mmlab · Feb 11, 2022 · a25f360 · a25f360
1 parent 5ae4609
commit a25f360
Show file tree

Hide file tree

Showing 88 changed files with 1,490 additions and 394 deletions.
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
@@ -44,6 +44,7 @@ jobs:
       - name: Install unittest dependencies
         run: |
           pip install -r requirements.txt
+          pip install -U numpy
       - name: Build and install
         run: rm -rf .eggs && pip install -e .
       - name: Run unittests and generate coverage report
@@ -85,6 +86,7 @@ jobs:
           python -V
           python -m pip install mmcv-full==${{matrix.mmcv}} -f https://download.openmmlab.com/mmcv/dist/cu102/${{matrix.torch_version}}/index.html
           python -m pip install -r requirements.txt
+          python -m pip install -U numpy
       - name: Build and install
         run: |
           rm -rf .eggs && python -m pip install -e .
@@ -128,6 +130,7 @@ jobs:
           python -V
           python -m pip install mmcv-full==${{matrix.mmcv}} -f https://download.openmmlab.com/mmcv/dist/cu111/${{matrix.torch_version}}/index.html
           python -m pip install -r requirements.txt
+          python -m pip install -U numpy
       - name: Build and install
         run: |
           rm -rf .eggs && python -m pip install -e .

diff --git a/.isort.cfg b/.isort.cfg
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -3,12 +3,8 @@ repos:
     rev: 4.0.1
     hooks:
       - id: flake8
-  - repo: https://github.com/asottile/seed-isort-config
-    rev: v2.2.0
-    hooks:
-      - id: seed-isort-config
-  - repo: https://github.com/timothycrosley/isort
-    rev: 4.3.21
+  - repo: https://github.com/PyCQA/isort
+    rev: 5.10.1
     hooks:
       - id: isort
   - repo: https://github.com/pre-commit/mirrors-yapf

diff --git a/configs/mmdet/detection/single-stage_ncnn_static-416x416.py b/configs/mmdet/detection/single-stage_ncnn_static-416x416.py
@@ -0,0 +1,4 @@
+_base_ = ['../_base_/base_static.py', '../../_base_/backends/ncnn.py']
+
+codebase_config = dict(model_type='ncnn_end2end')
+onnx_config = dict(output_names=['detection_output'], input_shape=[416, 416])
diff --git a/configs/mmseg/segmentation_pplnn_static-512x1024.py b/configs/mmseg/segmentation_pplnn_static-512x1024.py
@@ -0,0 +1,5 @@
+_base_ = ['./segmentation_static.py', '../_base_/backends/pplnn.py']
+
+onnx_config = dict(input_shape=[1024, 512])
+
+backend_config = dict(model_inputs=dict(opt_shape=[1, 3, 512, 1024]))
diff --git a/csrc/apis/c/detector.cpp b/csrc/apis/c/detector.cpp
@@ -155,7 +155,7 @@ MM_SDK_API void mmdeploy_detector_release_result(mm_detect_t* results, const int
   for (int i = 0; i < count; ++i) {
     for (int j = 0; j < result_count[i]; ++j, ++result_ptr) {
       if (result_ptr->mask) {
-        delete result_ptr->mask->data;
+        delete[] result_ptr->mask->data;
         delete result_ptr->mask;
       }
     }

diff --git a/csrc/backend_ops/torchscript/bind.cpp b/csrc/backend_ops/torchscript/bind.cpp
@@ -1,3 +1,4 @@
+// Copyright (c) OpenMMLab. All rights reserved.
 #include "torch/script.h"
 
 TORCH_LIBRARY(mmdeploy, m) {

diff --git a/csrc/backend_ops/torchscript/ops/modulated_deform_conv/modulated_deform_conv_cpu.cpp b/csrc/backend_ops/torchscript/ops/modulated_deform_conv/modulated_deform_conv_cpu.cpp
@@ -1,3 +1,4 @@
+// Copyright (c) OpenMMLab. All rights reserved.
 #include "modulated_deform_conv/modulated_deform_conv_cpu.h"
 
 #include "torch/script.h"

diff --git a/csrc/backend_ops/torchscript/ops/modulated_deform_conv/modulated_deform_conv_cuda.cu b/csrc/backend_ops/torchscript/ops/modulated_deform_conv/modulated_deform_conv_cuda.cu
@@ -1,3 +1,4 @@
+// Copyright (c) OpenMMLab. All rights reserved.
 #include "c10/cuda/CUDAStream.h"
 #include "modulated_deform_conv/modulated_deform_conv_cuda.cuh"
 #include "torch/script.h"

diff --git a/csrc/backend_ops/torchscript/optimizer/optimizer.cpp b/csrc/backend_ops/torchscript/optimizer/optimizer.cpp
@@ -1,7 +1,15 @@
+// Copyright (c) OpenMMLab. All rights reserved.
 #include "optimizer.h"
 
+#include <torch/csrc/jit/passes/canonicalize_graph_fuser_ops.h>
+#include <torch/csrc/jit/passes/common_subexpression_elimination.h>
+#include <torch/csrc/jit/passes/constant_pooling.h>
+#include <torch/csrc/jit/passes/constant_propagation.h>
+#include <torch/csrc/jit/passes/dead_code_elimination.h>
 #include <torch/csrc/jit/passes/freeze_module.h>
 #include <torch/csrc/jit/passes/frozen_graph_optimizations.h>
+#include <torch/csrc/jit/passes/peephole.h>
+#include <torch/csrc/jit/passes/remove_expands.h>
 
 #if TORCH_VERSION_MINOR >= 9
 #include <torch/csrc/jit/passes/frozen_conv_add_relu_fusion.h>
@@ -10,6 +18,15 @@
 #endif
 
 namespace mmdeploy {
+
+using torch::jit::Graph;
+const std::shared_ptr<Graph>& required_passes(const std::shared_ptr<Graph>& graph) {
+  RemoveExpands(graph);
+  CanonicalizeOps(graph);
+  EliminateDeadCode(graph);
+  return graph;
+}
+
 Module optimize_for_torchscript(const Module& model) {
   auto frozen_model = freeze_module(model);
   auto graph = frozen_model.get_method("forward").graph();
@@ -21,6 +38,12 @@ Module optimize_for_torchscript(const Module& model) {
   FrozenLinearTranspose(graph);
 #endif
 
+  graph = required_passes(graph);
+  EliminateCommonSubexpression(graph);
+  PeepholeOptimize(graph);
+  ConstantPropagation(graph);
+  ConstantPooling(graph);
+
   // TODO: add more custom passes
 
   return frozen_model;

diff --git a/csrc/backend_ops/torchscript/optimizer/optimizer.h b/csrc/backend_ops/torchscript/optimizer/optimizer.h
@@ -1,3 +1,4 @@
+// Copyright (c) OpenMMLab. All rights reserved.
 #include <torch/script.h>
 
 namespace mmdeploy {

diff --git a/csrc/codebase/common.h b/csrc/codebase/common.h
@@ -14,7 +14,7 @@ namespace mmdeploy {
 class Context {
  public:
   explicit Context(const Value& config) {
-    DEBUG("config: {}", cfg);
+    DEBUG("config: {}", config);
     device_ = config["context"]["device"].get<Device>();
     stream_ = config["context"]["stream"].get<Stream>();
   }

diff --git a/csrc/codebase/mmdet/object_detection.cpp b/csrc/codebase/mmdet/object_detection.cpp
@@ -107,8 +107,7 @@ Result<DetectorOutput> ResizeBBox::GetBBoxes(const Value& prep_res, const Tensor
             rect[3] - rect[1]);
       continue;
     }
-    DEBUG("remap left {}, top {}, right {}, bottom {}", rect.left, rect.top, rect.right,
-          rect.bottom);
+    DEBUG("remap left {}, top {}, right {}, bottom {}", rect[0], rect[1], rect[2], rect[3]);
     DetectorOutput::Detection det{};
     det.index = i;
     det.label_id = static_cast<int>(*labels_ptr);

diff --git a/demo/csrc/CMakeLists.txt b/demo/csrc/CMakeLists.txt
@@ -7,8 +7,8 @@ find_package(MMDeploy REQUIRED)
 
 function(add_example name)
   add_executable(${name} ${name}.cpp)
-  target_link_libraries(${name} ${MMDeploy_LIBS} opencv_imgcodecs
-          opencv_imgproc opencv_core)
+  target_link_libraries(${name} ${MMDeploy_LIBS} -Wl,--disable-new-dtags
+  opencv_imgcodecs opencv_imgproc opencv_core)
 endfunction()
 
 add_example(image_classification)

diff --git a/demo/csrc/object_detection.cpp b/demo/csrc/object_detection.cpp
@@ -37,12 +37,42 @@ int main(int argc, char *argv[]) {
     return 1;
   }
 
-  fprintf(stderr, "bbox_count=%d\n", *res_count);
+  fprintf(stdout, "bbox_count=%d\n", *res_count);
 
   for (int i = 0; i < *res_count; ++i) {
     const auto &box = bboxes[i].bbox;
-    fprintf(stderr, "box %d, left=%.2f, top=%.2f, right=%.2f, bottom=%.2f, label=%d, score=%.4f\n",
+    const auto &mask = bboxes[i].mask;
+
+    fprintf(stdout, "box %d, left=%.2f, top=%.2f, right=%.2f, bottom=%.2f, label=%d, score=%.4f\n",
             i, box.left, box.top, box.right, box.bottom, bboxes[i].label_id, bboxes[i].score);
+
+    // skip detections with invalid bbox size (bbox height or width < 1)
+    if ((box.right - box.left) < 1 || (box.bottom - box.top) < 1) {
+      continue;
+    }
+
+    // skip detections less than specified score threshold
+    if (bboxes[i].score < 0.1) {
+      continue;
+    }
+
+    // generate mask overlay if model exports masks
+    if (mask != nullptr) {
+      fprintf(stdout, "mask %d, height=%d, width=%d\n", i, mask->height, mask->width);
+
+      cv::Mat imgMask(mask->height, mask->width, CV_8UC1, &mask->data[0]);
+      auto x0 = std::max(std::floor(box.left) - 1, 0.f);
+      auto y0 = std::max(std::floor(box.top) - 1, 0.f);
+      cv::Rect roi((int)x0, (int)y0, mask->width, mask->height);
+
+      // split the RGB channels, overlay mask to a specific color channel
+      cv::Mat ch[3];
+      split(img, ch);
+      int col = 0;  // int col = i % 3;
+      cv::bitwise_or(imgMask, ch[col](roi), ch[col](roi));
+      merge(ch, 3, img);
+    }
+
     cv::rectangle(img, cv::Point{(int)box.left, (int)box.top},
                   cv::Point{(int)box.right, (int)box.bottom}, cv::Scalar{0, 255, 0});
   }

diff --git a/docker/CPU/Dockerfile b/docker/CPU/Dockerfile
@@ -0,0 +1,105 @@
+FROM openvino/ubuntu18_dev:2021.4.2
+ARG PYTHON_VERSION=3.7
+ARG TORCH_VERSION=1.8.0
+ARG TORCHVISION_VERSION=0.9.0
+ARG ONNXRUNTIME_VERSION=1.8.1
+ARG MMCV_VERSION=1.4.0
+ARG CMAKE_VERSION=3.20.0
+USER root
+RUN apt-get update && apt-get install -y --no-install-recommends \
+        ca-certificates \
+        libopencv-dev libspdlog-dev \
+        gnupg \
+        libssl-dev \
+        libprotobuf-dev protobuf-compiler \
+        build-essential \
+        libjpeg-dev \
+        libpng-dev \
+        ccache \
+        cmake \
+        gcc \
+        g++ \
+        git \
+        vim \
+        wget \
+        curl \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN curl -fsSL -v -o ~/miniconda.sh -O  https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh  && \
+    chmod +x ~/miniconda.sh && \
+    ~/miniconda.sh -b -p /opt/conda && \
+    rm ~/miniconda.sh && \
+    /opt/conda/bin/conda install -y python=${PYTHON_VERSION} conda-build pyyaml numpy ipython cython typing typing_extensions mkl mkl-include ninja && \
+    /opt/conda/bin/conda clean -ya
+
+### pytorch
+RUN /opt/conda/bin/pip install torch==${TORCH_VERSION}+cpu torchvision==${TORCHVISION_VERSION}+cpu -f https://download.pytorch.org/whl/cpu/torch_stable.html
+ENV PATH /opt/conda/bin:$PATH
+
+### install open-mim
+RUN /opt/conda/bin/pip install mmcv-full==${MMCV_VERSION} -f https://download.openmmlab.com/mmcv/dist/cpu/torch${TORCH_VERSION}/index.html
+
+WORKDIR /root/workspace
+
+### get onnxruntime
+RUN wget https://github.com/microsoft/onnxruntime/releases/download/v${ONNXRUNTIME_VERSION}/onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}.tgz \
+    && tar -zxvf onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}.tgz
+
+ENV ONNXRUNTIME_DIR=/root/workspace/onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}
+
+### update cmake to 20
+RUN wget https://github.com/Kitware/CMake/releases/download/v${CMAKE_VERSION}/cmake-${CMAKE_VERSION}.tar.gz &&\
+    tar -zxvf cmake-${CMAKE_VERSION}.tar.gz &&\
+    cd cmake-${CMAKE_VERSION} &&\
+    ./bootstrap &&\
+    make &&\
+    make install
+
+### install onnxruntme and openvino
+RUN /opt/conda/bin/pip install onnxruntime==${ONNXRUNTIME_VERSION} openvino-dev
+
+### build ncnn
+RUN git clone https://github.com/Tencent/ncnn.git &&\
+    cd ncnn &&\
+    export NCNN_DIR=$(pwd) &&\
+    git submodule update --init &&\
+    mkdir -p build && cd build &&\
+    cmake -DNCNN_VULKAN=OFF -DNCNN_SYSTEM_GLSLANG=ON -DNCNN_BUILD_EXAMPLES=ON -DNCNN_PYTHON=ON -DNCNN_BUILD_TOOLS=ON -DNCNN_BUILD_BENCHMARK=ON -DNCNN_BUILD_TESTS=ON .. &&\
+    make install &&\
+    cd /root/workspace/ncnn/python &&\
+    pip install -e .
+
+### install mmdeploy
+WORKDIR /root/workspace
+ARG VERSION
+RUN git clone https://github.com/open-mmlab/mmdeploy.git &&\
+    cd mmdeploy &&\
+    if [ -z ${VERSION} ] ; then echo "No MMDeploy version passed in, building on master" ; else git checkout tags/v${VERSION} -b tag_v${VERSION} ; fi &&\
+    git submodule update --init --recursive &&\
+    rm -rf build &&\
+    mkdir build &&\
+    cd build &&\
+    cmake -DMMDEPLOY_TARGET_BACKENDS=ncnn -Dncnn_DIR=/root/workspace/ncnn/build/install/lib/cmake/ncnn .. &&\
+    make -j$(nproc) &&\
+    cmake -DMMDEPLOY_TARGET_BACKENDS=ort .. &&\
+    make -j$(nproc) &&\
+    cd .. &&\
+    pip install -e .
+
+### build SDK
+ENV LD_LIBRARY_PATH="/root/workspace/mmdeploy/build/lib:/opt/intel/openvino/deployment_tools/ngraph/lib:/opt/intel/openvino/deployment_tools/inference_engine/lib/intel64:${LD_LIBRARY_PATH}"
+RUN cd mmdeploy && rm -rf build/CM* && mkdir -p build && cd build && cmake .. \
+    -DMMDEPLOY_BUILD_SDK=ON \
+    -DCMAKE_CXX_COMPILER=g++-7 \
+    -DONNXRUNTIME_DIR=${ONNXRUNTIME_DIR} \
+    -Dncnn_DIR=/root/workspace/ncnn/build/install/lib/cmake/ncnn \
+    -DInferenceEngine_DIR=/opt/intel/openvino/deployment_tools/inference_engine/share \
+    -DMMDEPLOY_TARGET_DEVICES=cpu \
+    -DMMDEPLOY_BUILD_SDK_PYTHON_API=ON \
+    -DMMDEPLOY_TARGET_BACKENDS="ort;ncnn;openvino" \
+    -DMMDEPLOY_CODEBASES=all &&\
+    cmake --build . -- -j$(nproc) && cmake --install . &&\
+    cd install/example  && mkdir -p build && cd build &&\
+    cmake -DMMDeploy_DIR=/root/workspace/mmdeploy/build/install/lib/cmake/MMDeploy .. &&\
+    cmake --build . && export SPDLOG_LEVEL=warn &&\
+    if [ -z ${VERSION} ] ; then echo "Built MMDeploy master for CPU devices successfully!" ; else echo "Built MMDeploy version v${VERSION} for CPU devices successfully!" ; fi
diff --git a/docker/GPU/Dockerfile b/docker/GPU/Dockerfile
@@ -0,0 +1,90 @@
+FROM nvcr.io/nvidia/tensorrt:21.04-py3
+
+ARG CUDA=10.2
+ARG PYTHON_VERSION=3.8
+ARG TORCH_VERSION=1.8.0
+ARG TORCHVISION_VERSION=0.9.0
+ARG ONNXRUNTIME_VERSION=1.8.1
+ARG MMCV_VERSION=1.4.0
+ARG CMAKE_VERSION=3.20.0
+ENV FORCE_CUDA="1"
+
+ENV DEBIAN_FRONTEND=noninteractive
+
+### update apt and install libs
+RUN apt-get update &&\
+    apt-get install -y vim libsm6 libxext6 libxrender-dev libgl1-mesa-glx git wget libssl-dev libopencv-dev libspdlog-dev --no-install-recommends &&\
+    rm -rf /var/lib/apt/lists/*
+
+RUN curl -fsSL -v -o ~/miniconda.sh -O  https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh  && \
+    chmod +x ~/miniconda.sh && \
+    ~/miniconda.sh -b -p /opt/conda && \
+    rm ~/miniconda.sh && \
+    /opt/conda/bin/conda install -y python=${PYTHON_VERSION} conda-build pyyaml numpy ipython cython typing typing_extensions mkl mkl-include ninja && \
+    /opt/conda/bin/conda clean -ya
+
+### pytorch
+RUN /opt/conda/bin/conda install pytorch==${TORCH_VERSION} torchvision==${TORCHVISION_VERSION} cudatoolkit=${CUDA} -c pytorch
+ENV PATH /opt/conda/bin:$PATH
+
+### install mmcv-full
+RUN /opt/conda/bin/pip install mmcv-full==${MMCV_VERSION} -f https://download.openmmlab.com/mmcv/dist/cu${CUDA//./}/torch${TORCH_VERSION}/index.html
+
+WORKDIR /root/workspace
+### get onnxruntime
+RUN wget https://github.com/microsoft/onnxruntime/releases/download/v${ONNXRUNTIME_VERSION}/onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}.tgz \
+    && tar -zxvf onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}.tgz &&\
+    pip install onnxruntime-gpu==${ONNXRUNTIME_VERSION}
+
+### cp trt from pip to conda
+RUN cp -r /usr/local/lib/python${PYTHON_VERSION}/dist-packages/tensorrt* /opt/conda/lib/python${PYTHON_VERSION}/site-packages/
+
+### update cmake
+RUN  wget https://github.com/Kitware/CMake/releases/download/v${CMAKE_VERSION}/cmake-${CMAKE_VERSION}.tar.gz &&\
+    tar -zxvf cmake-${CMAKE_VERSION}.tar.gz &&\
+    cd cmake-${CMAKE_VERSION} &&\
+    ./bootstrap &&\
+    make &&\
+    make install
+
+### install mmdeploy
+ENV ONNXRUNTIME_DIR=/root/workspace/onnxruntime-linux-x64-${ONNXRUNTIME_VERSION}
+ENV TENSORRT_DIR=/workspace/tensorrt
+ARG VERSION
+RUN git clone https://github.com/open-mmlab/mmdeploy &&\
+    cd mmdeploy &&\
+    if [ -z ${VERSION} ] ; then echo "No MMDeploy version passed in, building on master" ; else git checkout tags/v${VERSION} -b tag_v${VERSION} ; fi &&\
+    git submodule update --init --recursive &&\
+    rm -rf build &&\
+    mkdir build &&\
+    cd build &&\
+    cmake -DMMDEPLOY_TARGET_BACKENDS=ort .. &&\
+    make -j$(nproc) &&\
+    cmake -DMMDEPLOY_TARGET_BACKENDS=trt .. &&\
+    make -j$(nproc) &&\
+    cd .. &&\
+    pip install -e .
+
+### build sdk
+RUN git clone https://github.com/openppl-public/ppl.cv.git &&\
+    cd ppl.cv &&\
+    ./build.sh cuda
+RUN cd /root/workspace/mmdeploy &&\
+    rm -rf build/CM* &&\
+    mkdir -p build && cd build &&\
+    cmake .. \
+        -DMMDEPLOY_BUILD_SDK=ON \
+        -DCMAKE_CXX_COMPILER=g++ \
+        -Dpplcv_DIR=/root/workspace/ppl.cv/cuda-build/install/lib/cmake/ppl \
+        -DTENSORRT_DIR=${TENSORRT_DIR} \
+        -DMMDEPLOY_BUILD_SDK_PYTHON_API=ON \
+        -DMMDEPLOY_TARGET_DEVICES="cuda;cpu" \
+        -DMMDEPLOY_TARGET_BACKENDS="trt" \
+        -DMMDEPLOY_CODEBASES=all &&\
+    cmake --build . -- -j$(nproc) && cmake --install . &&\
+    cd install/example  && mkdir -p build && cd build &&\
+    cmake -DMMDeploy_DIR=/root/workspace/mmdeploy/build/install/lib/cmake/MMDeploy .. &&\
+    cmake --build . && export SPDLOG_LEVEL=warn &&\
+    if [ -z ${VERSION} ] ; then echo "Built MMDeploy master for GPU devices successfully!" ; else echo "Built MMDeploy version v${VERSION} for GPU devices successfully!" ; fi
+
+ENV LD_LIBRARY_PATH="/root/workspace/mmdeploy/build/lib:${LD_LIBRARY_PATH}"