tensorrt_starter
This repository give a guidline to learn CUDA and TensorRT from the beginning.
파일 탐색기
최종 버전 다운로드 (.zip)- PULL_REQUEST_TEMPLATE.md
- fisher.fish
- __z.fish
- __z_add.fish
- __z_clean.fish
- __z_complete.fish
- fish_prompt.fish
- fish_user_key_bindings.fish
- fisher.fish
- peco_change_directory.fish
- peco_select_history.fish
- config.fish
- fish_plugins
- fish_variables
- clangd.json
- gopls.json
- pyright.json
- cpp.lua
- py.lua
- init.lua
- autocommands.lua
- chatgpt.lua
- keymaps.lua
- nvimtree.lua
- outlines.lua
- plugins.lua
- signature.lua
- telescope.lua
- treesitter.lua
- config.lua
- .tmux.conf
- .tmux.conf.osx
- .tmux.powerline.conf
- build-docker.sh
- run-docker.sh
- dockerfile
- README.md
- Makefile.config
- print_index.cu
- Makefile
- Makefile.config
- cat.png
- deer.png
- eagle.png
- fox.png
- tiny-cat.png
- unknown.png
- wolf.png
- preprocess.hpp
- timer.hpp
- utils.hpp
- main.cpp
- preprocess.cpp
- preprocess.cu
- timer.cpp
- utils.cpp
- Makefile
- Makefile.config
- cat.png
- deer.png
- eagle.png
- fox.png
- tiny-cat.png
- unknown.png
- wolf.png
- preprocess.hpp
- timer.hpp
- utils.hpp
- main.cpp
- preprocess.cpp
- preprocess.cu
- timer.cpp
- utils.cpp
- Makefile
- Makefile.config
- cat.png
- deer.png
- eagle.png
- fox.png
- tiny-cat.png
- unknown.png
- wolf.png
- preprocess.hpp
- timer.hpp
- utils.hpp
- main.cpp
- preprocess.cpp
- preprocess.cu
- timer.cpp
- utils.cpp
- Makefile
- Makefile.config
- main.cpp
- print_index.cu
- print_index.hpp
- utils.hpp
- Makefile
- Makefile.config
- main.cpp
- matmul.hpp
- matmul_cpu.cpp
- matmul_gpu_basic.cu
- timer.hpp
- utils.cpp
- utils.hpp
- Makefile
- Makefile.config
- main.cpp
- matmul.hpp
- matmul_cpu.cpp
- matmul_gpu_basic.cu
- timer.hpp
- utils.cpp
- utils.hpp
- Makefile
- Makefile.config
- main.cpp
- utils.cpp
- utils.hpp
- Makefile
- Makefile.config
- main.cpp
- matmul.hpp
- matmul_cpu.cpp
- matmul_gpu_basic.cu
- timer.hpp
- utils.cpp
- utils.hpp
- Makefile
- Makefile.config
- main.cpp
- matmul.hpp
- matmul_cpu.cpp
- matmul_gpu_basic.cu
- matmul_gpu_shared.cu
- timer.cpp
- timer.hpp
- utils.cpp
- utils.hpp
- Makefile
- Makefile.config
- main.cpp
- matmul.hpp
- matmul_cpu.cpp
- matmul_gpu_bank_conflict.cu
- matmul_gpu_bank_conflict_pad.cu
- matmul_gpu_basic.cu
- matmul_gpu_shared.cu
- timer.cpp
- timer.hpp
- utils.cpp
- utils.hpp
- Makefile
- Makefile.config
- gelu.cu
- gelu.hpp
- main.cpp
- stream.cu
- stream.hpp
- timer.cpp
- timer.hpp
- utils.cpp
- utils.hpp
- Makefile
- example.py
- example_dynamic_shape.py
- example_two_head.py
- load_torchvision.py
- sample_cbr.py
- sample_reshape.py
- create_onnx_convnet.py
- create_onnx_linear.py
- parse_onnx_cbr.py
- parse_onnx_convnet.py
- parse_onnx_linear.py
- parser.py
- sample_asinh.py
- sample_asinh_register.py
- sample_asinh_register2.py
- sample_custom_op_autograd.py
- sample_custom_op_autograd_register.py
- sample_deformable_conv.py
- sample_deformable_conv_register.py
- gs_create_complicate_graph.py
- gs_create_conv.py
- gs_create_subgraph.py
- gs_remove_node.py
- gs_replace_LN.py
- gs_replace_op.py
- export.py
- symbolic_opset12-custom.py
- .gitignore
- load_torchvision.py
- sample_cbr.py
- sample_reshape.py
- build.sh
- infer.sh
- profile.sh
- README.md
- Makefile
- sampleOnnxMNIST_CN.cpp
- README.md
- settings.json
- Makefile.config
- main.cpp
- model.cpp
- model.hpp
- utils.cpp
- utils.hpp
- generate_onnx.py
- build.sh
- infer.sh
- profile.sh
- Makefile
- Makefile.config
- main.cpp
- model.cpp
- model.hpp
- utils.cpp
- utils.hpp
- generate_onnx.py
- build.sh
- infer.sh
- profile.sh
- Makefile
- Makefile.config
- main.cpp
- model.cpp
- model.hpp
- utils.cpp
- utils.hpp
- generate_onnx.py
- build.sh
- infer.sh
- profile.sh
- Makefile
- Makefile.config
- main.cpp
- model.cpp
- model.hpp
- utils.cpp
- utils.hpp
- export_batchNorm.py
- export_cbr.py
- export_cbr_dynamic.py
- export_concat.py
- export_conv.py
- export_deconvolution.py
- export_elementwise.py
- export_linear.py
- export_permute.py
- export_pooling.py
- export_reduce.py
- export_reshape.py
- export_slice.py
- export_unsample.py
- build.sh
- infer.sh
- profile.sh
- Makefile
- Makefile.config
- sample_c2f.weights
- sample_cbr.weights
- sample_convBNSiLU.weights
- sample_resBlock.weights
- main.cpp
- model.cpp
- model.hpp
- network.cpp
- network.hpp
- parser.cpp
- utils.cpp
- utils.hpp
- export_batchNorm.py
- export_c2f.py
- export_cbr.py
- export_cbr_dynamic.py
- export_conv.py
- export_convBNSiLU.py
- export_linear.py
- export_permute.py
- export_resBlock.py
- export_reshape.py
- build.sh
- infer.sh
- profile.sh
- Makefile
- Makefile.config
- custom-leakyReLU-kernel.cu
- custom-leakyReLU-plugin.cpp
- custom-leakyReLU-plugin.hpp
- custom-scalar-kernel.cu
- custom-scalar-plugin.cpp
- custom-scalar-plugin.hpp
- main.cpp
- model.cpp
- model.hpp
- network.cpp
- network.hpp
- parser.cpp
- test.cpp
- utils.cpp
- utils.hpp
- plugin_customLeakyReLU.py
- plugin_customScalar.py
- build.sh
- infer.sh
- profile.sh
- Makefile
- Makefile.config
- custom-plugin.log
- custom-leakyReLU-kernel.cu
- custom-leakyReLU-plugin.cpp
- custom-leakyReLU-plugin.hpp
- custom-scalar-kernel.cu
- custom-scalar-plugin.cpp
- custom-scalar-plugin.hpp
- utils.cpp
- utils.hpp
- logger.py
- plugin_customLeakyReLU.py
- plugin_customScalar.py
- test.log
- trt_model.py
- unit_test_customLeakyReLU.py
- unit_test_customScalar.py
- build.sh
- infer.sh
- profile.sh
- Makefile
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- Makefile.config
- fox-cvtcolor.png
- fox-mat-at.png
- fox-mat-data.png
- fox-mat-iterator.png
- fox-pointer.png
- fox.png
- imagenet_labels.hpp
- model.hpp
- preprocess.hpp
- timer.hpp
- utils.hpp
- build.sh
- infer.sh
- profile.sh
- main.cpp
- model.cpp
- preprocess.cpp
- timer.cpp
- utils.cpp
- export_pretained.py
- .clangd
- Makefile
- README.md
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- Makefile.config
- cat.png
- eagle.png
- fox.png
- gazelle.png
- tiny-cat.png
- unknown.png
- wolf.png
- imagenet_labels.hpp
- model.hpp
- timer.hpp
- utils.hpp
- main.cpp
- model.cpp
- timer.cpp
- utils.cpp
- export_pretained.py
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- Makefile.config
- cat.png
- eagle.png
- fox.png
- gazelle.png
- tiny-cat.png
- unknown.png
- wolf.png
- imagenet_labels.hpp
- trt_classifier.hpp
- trt_logger.hpp
- trt_model.hpp
- trt_preprocess.hpp
- trt_timer.hpp
- trt_worker.hpp
- utils.hpp
- main.cpp
- trt_classifier.cpp
- trt_logger.cpp
- trt_model.cpp
- trt_preprocess.cpp
- trt_preprocess.cu
- trt_timer.cpp
- trt_worker.cpp
- utils.cpp
- export_pretained.py
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- calibration_list.txt
- calibration_list_imagenet.txt
- Makefile.config
- cat.png
- eagle.png
- fox.png
- gazelle.png
- tiny-cat.png
- unknown.png
- wolf.png
- imagenet_labels.hpp
- trt_calibrator.hpp
- trt_classifier.hpp
- trt_logger.hpp
- trt_model.hpp
- trt_preprocess.hpp
- trt_timer.hpp
- trt_worker.hpp
- utils.hpp
- main.cpp
- trt_calibrator.cpp
- trt_classifier.cpp
- trt_logger.cpp
- trt_model.cpp
- trt_preprocess.cpp
- trt_preprocess.cu
- trt_timer.cpp
- trt_worker.cpp
- utils.cpp
- export_pretained.py
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- calibration_list.txt
- calibration_list_imagenet.txt
- Makefile.config
- cat.png
- eagle.png
- fox.png
- gazelle.png
- tiny-cat.png
- unknown.png
- wolf.png
- imagenet_labels.hpp
- trt_calibrator.hpp
- trt_classifier.hpp
- trt_logger.hpp
- trt_model.hpp
- trt_preprocess.hpp
- trt_timer.hpp
- trt_worker.hpp
- utils.hpp
- mobilenetV2.onnx.engine.build.log
- mobilenetV2.onnx.engine.build.metadata.json
- resnet50-int8.engine.graph.json
- resnet50-int8.engine.graph.json.gv
- resnet50-int8.engine.graph.json.svg
- resnet50-int8.engine.profile.json
- resnet50-int8.engine.profile.log
- resnet50-int8.engine.profile.metadata.json
- resnet50-int8.engine.timing.json
- main.cpp
- trt_calibrator.cpp
- trt_classifier.cpp
- trt_logger.cpp
- trt_model.cpp
- trt_preprocess.cpp
- trt_preprocess.cu
- trt_timer.cpp
- trt_worker.cpp
- utils.cpp
- config_gpu.py
- draw_engine.py
- export_pretained.py
- parse_trtexec_log.py
- process_engine.py
- timing.cache
- timing.cache.lock
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- Makefile.config
- data_cpp.npy
- data_cpp.npz
- data_python.npz
- trt_logger.hpp
- utils.hpp
- info.log
- main.cpp
- trt_logger.cpp
- utils.cpp
- create_data.py
- logger.py
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- Makefile.config
- cat.png
- deer.png
- eagle.png
- fox.png
- tiny-cat.png
- unknown.png
- wolf.png
- preprocess.hpp
- timer.hpp
- utils.hpp
- main.cpp
- preprocess.cpp
- preprocess.cu
- timer.cpp
- utils.cpp
- .clangd
- Makefile
- README.md
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- calibration_list.txt
- calibration_list_coco.txt
- Makefile.config
- airport.jpg
- bedroom.jpg
- car.jpg
- crossroad.jpg
- crowd.jpg
- coco_labels.hpp
- imagenet_labels.hpp
- trt_calibrator.hpp
- trt_classifier.hpp
- trt_detector.hpp
- trt_logger.hpp
- trt_model.hpp
- trt_preprocess.hpp
- trt_timer.hpp
- trt_worker.hpp
- utils.hpp
- main.cpp
- trt_calibrator.cpp
- trt_classifier.cpp
- trt_detector.cpp
- trt_logger.cpp
- trt_model.cpp
- trt_preprocess.cpp
- trt_preprocess.cu
- trt_timer.cpp
- trt_worker.cpp
- utils.cpp
- export_cbr_dynamic.py
- export_pretained.py
- export_yolo.py
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- calibration_list.txt
- calibration_list_coco.txt
- Makefile.config
- airport-detect-fp16.png
- airport-detect-fp32.png
- airport-detect-int8.png
- bedroom-detect-fp16.png
- bedroom-detect-fp32.png
- bedroom-detect-int8.png
- car-detect-fp16.png
- car-detect-fp32.png
- car-detect-int8.png
- crossroad-detect-fp16.png
- crossroad-detect-fp32.png
- crossroad-detect-int8.png
- airport.jpg
- bedroom.jpg
- car.jpg
- crossroad.jpg
- crowd.jpg
- coco_labels.hpp
- imagenet_labels.hpp
- trt_calibrator.hpp
- trt_classifier.hpp
- trt_detector.hpp
- trt_logger.hpp
- trt_model.hpp
- trt_preprocess.hpp
- trt_timer.hpp
- trt_worker.hpp
- utils.hpp
- main.cpp
- trt_calibrator.cpp
- trt_classifier.cpp
- trt_detector.cpp
- trt_logger.cpp
- trt_model.cpp
- trt_preprocess.cpp
- trt_preprocess.cu
- trt_timer.cpp
- trt_worker.cpp
- utils.cpp
- config_gpu.py
- draw_engine.py
- export_cbr_dynamic.py
- export_pretained.py
- parse_trtexec_log.py
- process_engine.py
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- c_cpp_properties.json
- launch.json
- settings.json
- tasks.json
- calibration_list.txt
- calibration_list_coco.txt
- Makefile.config
- airport.jpg
- bedroom.jpg
- car.jpg
- crossroad.jpg
- crowd.jpg
- coco_labels.hpp
- imagenet_labels.hpp
- trt_calibrator.hpp
- trt_classifier.hpp
- trt_detector.hpp
- trt_logger.hpp
- trt_model.hpp
- trt_multitask.hpp
- trt_preprocess.hpp
- trt_timer.hpp
- trt_worker.hpp
- utils.hpp
- main.cpp
- trt_calibrator.cpp
- trt_classifier.cpp
- trt_detector.cpp
- trt_logger.cpp
- trt_model.cpp
- trt_multitask.cpp
- trt_preprocess.cpp
- trt_preprocess.cu
- trt_timer.cpp
- trt_worker.cpp
- utils.cpp
- export_cbr_dynamic.py
- export_pretained.py
- export_yolo.py
- build.sh
- infer.sh
- profile.sh
- .clangd
- Makefile
- README.md
- Makefile.config
- .gitignore
- .gitmodules
- README.md
# 설치 가이드
1. 코드 내려받기
git clone https://github.com/kalfazed/tensorrt_starter
깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.
cd tensorrt_starter
방금 내려받은 프로젝트 폴더 안으로 이동합니다.
2. Docker
쉬움 추천사전 준비물
- Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
- Docker Desktop 컨테이너를 빌드하고 실행하려면 필요합니다. 설치 후 실행해서 백그라운드에 켜두세요.
⚠️ 이 프로젝트는 규모가 큰 저장소라, 이 방법이 실제 핵심 제품이 아니라 내부 하위 패키지를 가리키는 것일 수 있습니다. README 전체를 함께 확인해보세요.
docker build -f chapter1-build-environment/1.0-build-environment/dockerfile -t tensorrt-starter .
Dockerfile을 기반으로 실행 가능한 이미지를 빌드합니다.
docker run -p 8080:80 tensorrt-starter
빌드된 이미지를 실제 컨테이너로 실행합니다.
터미널에 docker compose ps 를 입력해 컨테이너들이 Up 상태인지 확인하세요. README에 포트 번호가 적혀있다면 브라우저에서 http://localhost:포트번호 로 접속해보세요.
3. Make
보통사전 준비물
- Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
- Make Linux/macOS는 보통 기본 설치되어 있습니다. Windows는 별도 설치(예: MSYS2, WSL)가 필요합니다.
⚠️ 이 프로젝트는 규모가 큰 저장소라, 이 방법이 실제 핵심 제품이 아니라 내부 하위 패키지를 가리키는 것일 수 있습니다. README 전체를 함께 확인해보세요.
cd chapter2-cuda-programming/2.1-dim_and_index
이 프로젝트의 관련 파일이 하위 폴더 안에 있어서, 먼저 그 폴더로 이동합니다.
make
생성된 빌드 설정을 바탕으로 실제 컴파일을 진행해 실행 파일을 만듭니다.
에러 없이 끝나면 성공입니다. 생성된 실행 파일을 직접 실행해보세요.
// repository documentation
Was this content helpful?
(0 ratings)
