目录
风之痕
诗号： 昂首千丘远，啸傲风间；堪寻敌手共论剑，高处不胜寒。
魔界传说中的剑术高手，无人能出其右。平生追求剑术极致，崇尚自由不受拘束，创造出风之痕、魔流剑二种截然不同的剑法，一为冷静快意、一为狂野疯狂。与剑痞忆秋年论交，为知己、也为堪称敌手的朋友！魔吞十二宫任务已了，与徒弟退隐多时，因故前往异域镜界，却意外成了特殊融合姿态，有了新的突破契机。
一、概述
ls-ml-toolkit 是一个专门为 Label Studio 设计的机器学习工具包，它最大的特点就是能帮你把标注好的数据一键转换并开始训练，省去了很多手动整理的麻烦。
二、安装
2.1 启用python虚拟环境
进入虚拟机安装路径，启动虚拟机
bash
zero@PC-LJM:~$ cd label-studio-project/
zero@PC-LJM:~/label-studio-project$ cd label-studio-env/
zero@PC-LJM:~/label-studio-project/label-studio-env$ source bin/activate2.2 安装工具
使用pip直接安装
bash
pip install ls-ml-toolkit安装信息，会安装很多配套的依赖。
bash
(label-studio-env) zero@PC-LJM:~/label-studio-project/label-studio-env$ pip install ls-ml-toolkit
Collecting ls-ml-toolkit
Downloading ls_ml_toolkit-1.0.2-py3-none-any.whl.metadata (15 kB)
Collecting ultralytics>=8.0.0 (from ls-ml-toolkit)
Downloading ultralytics-8.4.95-py3-none-any.whl.metadata (42 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 42.1/42.1 kB 397.2 kB/s eta 0:00:00
Collecting onnx>=1.15.0 (from ls-ml-toolkit)
Downloading onnx-1.22.0-cp312-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (8.5 kB)
Collecting onnxruntime>=1.16.0 (from ls-ml-toolkit)
Downloading onnxruntime-1.27.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (5.4 kB)
Requirement already satisfied: tqdm>=4.65.0 in ./lib/python3.12/site-packages (from ls-ml-toolkit) (4.68.4)
Requirement already satisfied: requests>=2.31.0 in ./lib/python3.12/site-packages (from ls-ml-toolkit) (2.32.5)
Requirement already satisfied: PyYAML>=6.0.0 in ./lib/python3.12/site-packages (from ls-ml-toolkit) (6.0.3)
Requirement already satisfied: boto3>=1.34.0 in ./lib/python3.12/site-packages (from ls-ml-toolkit) (1.43.45)
Requirement already satisfied: botocore>=1.34.0 in ./lib/python3.12/site-packages (from ls-ml-toolkit) (1.43.45)
Collecting torch>=2.0.0 (from ls-ml-toolkit)
Downloading torch-2.13.0-cp312-cp312-manylinux_2_28_x86_64.whl.metadata (38 kB)
Collecting torchvision>=0.15.0 (from ls-ml-toolkit)
Downloading torchvision-0.28.0-cp312-cp312-manylinux_2_28_x86_64.whl.metadata (5.6 kB)
Requirement already satisfied: jmespath<2.0.0,>=0.7.1 in ./lib/python3.12/site-packages (from boto3>=1.34.0->ls-ml-toolkit) (1.1.0)
Requirement already satisfied: s3transfer<0.20.0,>=0.19.0 in ./lib/python3.12/site-packages (from boto3>=1.34.0->ls-ml-toolkit) (0.19.0)
Requirement already satisfied: python-dateutil<3.0.0,>=2.1 in ./lib/python3.12/site-packages (from botocore>=1.34.0->ls-ml-toolkit) (2.9.0.post0)
Requirement already satisfied: urllib3!=2.2.0,<3,>=1.25.4 in ./lib/python3.12/site-packages (from botocore>=1.34.0->ls-ml-toolkit) (2.7.0)
Requirement already satisfied: numpy>=1.23.2 in ./lib/python3.12/site-packages (from onnx>=1.15.0->ls-ml-toolkit) (2.5.1)
Requirement already satisfied: protobuf>=4.25.1 in ./lib/python3.12/site-packages (from onnx>=1.15.0->ls-ml-toolkit) (7.35.1)
Requirement already satisfied: typing_extensions>=4.15.0 in ./lib/python3.12/site-packages (from onnx>=1.15.0->ls-ml-toolkit) (4.16.0)
Collecting ml_dtypes>=0.5.4 (from onnx>=1.15.0->ls-ml-toolkit)
Downloading ml_dtypes-0.5.4-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (8.9 kB)
Collecting flatbuffers (from onnxruntime>=1.16.0->ls-ml-toolkit)
Downloading flatbuffers-25.12.19-py2.py3-none-any.whl.metadata (1.0 kB)
Requirement already satisfied: packaging in ./lib/python3.12/site-packages (from onnxruntime>=1.16.0->ls-ml-toolkit) (26.2)
Requirement already satisfied: charset_normalizer<4,>=2 in ./lib/python3.12/site-packages (from requests>=2.31.0->ls-ml-toolkit) (3.4.9)
Requirement already satisfied: idna<4,>=2.5 in ./lib/python3.12/site-packages (from requests>=2.31.0->ls-ml-toolkit) (3.18)
Requirement already satisfied: certifi>=2017.4.17 in ./lib/python3.12/site-packages (from requests>=2.31.0->ls-ml-toolkit) (2026.6.17)
Requirement already satisfied: filelock in ./lib/python3.12/site-packages (from torch>=2.0.0->ls-ml-toolkit) (3.29.7)
Requirement already satisfied: setuptools>=77.0.3 in ./lib/python3.12/site-packages (from torch>=2.0.0->ls-ml-toolkit) (83.0.0)
Collecting sympy>=1.13.3 (from torch>=2.0.0->ls-ml-toolkit)
Downloading sympy-1.14.0-py3-none-any.whl.metadata (12 kB)
Collecting networkx>=2.5.1 (from torch>=2.0.0->ls-ml-toolkit)
Downloading networkx-3.6.1-py3-none-any.whl.metadata (6.8 kB)
Requirement already satisfied: jinja2 in ./lib/python3.12/site-packages (from torch>=2.0.0->ls-ml-toolkit) (3.1.6)
Collecting fsspec>=0.8.5 (from torch>=2.0.0->ls-ml-toolkit)
Downloading fsspec-2026.6.0-py3-none-any.whl.metadata (10 kB)
Collecting cuda-toolkit==13.0.3 (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading cuda_toolkit-13.0.3.0-py2.py3-none-any.whl.metadata (17 kB)
Collecting cuda-bindings<14,>=13.0.3 (from torch>=2.0.0->ls-ml-toolkit)
Downloading cuda_bindings-13.3.1-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl.metadata (2.5 kB)
Collecting nvidia-cudnn-cu13==9.20.0.48 (from torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cudnn_cu13-9.20.0.48-py3-none-manylinux_2_27_x86_64.whl.metadata (1.9 kB)
Collecting nvidia-cusparselt-cu13==0.8.1 (from torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cusparselt_cu13-0.8.1-py3-none-manylinux2014_x86_64.whl.metadata (12 kB)
Collecting nvidia-nccl-cu13==2.29.7 (from torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_nccl_cu13-2.29.7-py3-none-manylinux_2_18_x86_64.whl.metadata (2.1 kB)
Collecting nvidia-nvshmem-cu13==3.4.5 (from torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_nvshmem_cu13-3.4.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (2.1 kB)
Collecting triton==3.7.1 (from torch>=2.0.0->ls-ml-toolkit)
Downloading triton-3.7.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (1.7 kB)
Collecting nvidia-cublas==13.1.1.3.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cublas-13.1.1.3-py3-none-manylinux_2_27_x86_64.whl.metadata (1.8 kB)
Collecting nvidia-cuda-nvrtc==13.0.88.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cuda_nvrtc-13.0.88-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl.metadata (1.7 kB)
Collecting nvidia-cuda-runtime==13.0.96.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cuda_runtime-13.0.96-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.7 kB)
Collecting nvidia-cufft==12.0.0.61.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cufft-12.0.0.61-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.8 kB)
Collecting nvidia-nvjitlink<14,>=13.0.88 (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_nvjitlink-13.3.33-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl.metadata (1.8 kB)
Collecting nvidia-cufile==1.15.1.6.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cufile-1.15.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.7 kB)
Collecting nvidia-cuda-cupti==13.0.85.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cuda_cupti-13.0.85-py3-none-manylinux_2_25_x86_64.whl.metadata (1.7 kB)
Collecting nvidia-curand==10.4.0.35.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_curand-10.4.0.35-py3-none-manylinux_2_27_x86_64.whl.metadata (1.7 kB)
Collecting nvidia-cusolver==12.0.4.66.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cusolver-12.0.4.66-py3-none-manylinux_2_27_x86_64.whl.metadata (1.8 kB)
Collecting nvidia-cusparse==12.6.3.3.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_cusparse-12.6.3.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.8 kB)
Collecting nvidia-nvtx==13.0.85.* (from cuda-toolkit[cublas,cudart,cufft,cufile,cupti,curand,cusolver,cusparse,nvjitlink,nvrtc,nvtx]==13.0.3; platform_system == "Linux"->torch>=2.0.0->ls-ml-toolkit)
Downloading nvidia_nvtx-13.0.85-py3-none-manylinux1_x86_64.manylinux_2_5_x86_64.whl.metadata (1.8 kB)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in ./lib/python3.12/site-packages (from torchvision>=0.15.0->ls-ml-toolkit) (12.3.0)
Collecting matplotlib>=3.3.0 (from ultralytics>=8.0.0->ls-ml-toolkit)
Downloading matplotlib-3.11.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (80 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 80.3/80.3 kB 1.2 MB/s eta 0:00:00
Collecting opencv-python!=4.13.0.90,>=4.6.0 (from ultralytics>=8.0.0->ls-ml-toolkit)
Downloading opencv_python-5.0.0.93-cp37-abi3-manylinux_2_28_x86_64.whl.metadata (19 kB)
Collecting psutil>=5.8.0 (from ultralytics>=8.0.0->ls-ml-toolkit)
Downloading psutil-7.2.2-cp36-abi3-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl.metadata (22 kB)
Collecting polars>=0.20.0 (from ultralytics>=8.0.0->ls-ml-toolkit)
Downloading polars-1.42.1-py3-none-any.whl.metadata (10 kB)
Collecting nvidia-ml-py>=12.0.0 (from ultralytics>=8.0.0->ls-ml-toolkit)
Downloading nvidia_ml_py-13.610.43-py3-none-any.whl.metadata (9.7 kB)
Collecting ultralytics-thop>=2.0.18 (from ultralytics>=8.0.0->ls-ml-toolkit)
Downloading ultralytics_thop-2.0.20-py3-none-any.whl.metadata (14 kB)
Collecting cuda-pathfinder>=1.4.2 (from cuda-bindings<14,>=13.0.3->torch>=2.0.0->ls-ml-toolkit)
Downloading cuda_pathfinder-1.5.6-py3-none-any.whl.metadata (1.9 kB)
Collecting contourpy>=1.0.1 (from matplotlib>=3.3.0->ultralytics>=8.0.0->ls-ml-toolkit)
Downloading contourpy-1.3.3-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (5.5 kB)
Collecting cycler>=0.10 (from matplotlib>=3.3.0->ultralytics>=8.0.0->ls-ml-toolkit)
Downloading cycler-0.12.1-py3-none-any.whl.metadata (3.8 kB)
Collecting fonttools>=4.22.0 (from matplotlib>=3.3.0->ultralytics>=8.0.0->ls-ml-toolkit)
Downloading fonttools-4.63.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (118 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 118.7/118.7 kB 1.2 MB/s eta 0:00:00
Collecting kiwisolver>=1.3.1 (from matplotlib>=3.3.0->ultralytics>=8.0.0->ls-ml-toolkit)
Downloading kiwisolver-1.5.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (5.1 kB)
Collecting pyparsing>=3 (from matplotlib>=3.3.0->ultralytics>=8.0.0->ls-ml-toolkit)
Downloading pyparsing-3.3.2-py3-none-any.whl.metadata (5.8 kB)
Collecting polars-runtime-32==1.42.1 (from polars>=0.20.0->ultralytics>=8.0.0->ls-ml-toolkit)
Downloading polars_runtime_32-1.42.1-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (1.5 kB)
Requirement already satisfied: six>=1.5 in ./lib/python3.12/site-packages (from python-dateutil<3.0.0,>=2.1->botocore>=1.34.0->ls-ml-toolkit) (1.17.0)
Collecting mpmath<1.4,>=1.1.0 (from sympy>=1.13.3->torch>=2.0.0->ls-ml-toolkit)
Downloading mpmath-1.3.0-py3-none-any.whl.metadata (8.6 kB)
Requirement already satisfied: MarkupSafe>=2.0 in ./lib/python3.12/site-packages (from jinja2->torch>=2.0.0->ls-ml-toolkit) (3.0.3)
Downloading ls_ml_toolkit-1.0.2-py3-none-any.whl (25 kB)
Downloading onnx-1.22.0-cp312-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (19.1 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 19.1/19.1 MB 3.2 MB/s eta 0:00:00
Downloading onnxruntime-1.27.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (18.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 18.7/18.7 MB 2.9 MB/s eta 0:00:00
Downloading torch-2.13.0-cp312-cp312-manylinux_2_28_x86_64.whl (526.6 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 526.6/526.6 MB 5.7 MB/s eta 0:00:00
Downloading cuda_toolkit-13.0.3.0-py2.py3-none-any.whl (2.5 kB)
Downloading nvidia_cudnn_cu13-9.20.0.48-py3-none-manylinux_2_27_x86_64.whl (366.2 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 366.2/366.2 MB 6.0 MB/s eta 0:00:00
Downloading nvidia_cusparselt_cu13-0.8.1-py3-none-manylinux2014_x86_64.whl (170.1 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 170.1/170.1 MB 8.3 MB/s eta 0:00:00
Downloading nvidia_nccl_cu13-2.29.7-py3-none-manylinux_2_18_x86_64.whl (206.0 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 206.0/206.0 MB 7.6 MB/s eta 0:00:00
Downloading nvidia_nvshmem_cu13-3.4.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (60.4 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 60.4/60.4 MB 10.4 MB/s eta 0:00:00
Downloading triton-3.7.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (197.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 197.7/197.7 MB 8.0 MB/s eta 0:00:00
Downloading nvidia_cublas-13.1.1.3-py3-none-manylinux_2_27_x86_64.whl (423.1 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 423.1/423.1 MB 7.6 MB/s eta 0:00:00
Downloading nvidia_cuda_cupti-13.0.85-py3-none-manylinux_2_25_x86_64.whl (10.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 10.7/10.7 MB 4.7 MB/s eta 0:00:00
Downloading nvidia_cuda_nvrtc-13.0.88-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (90.2 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 90.2/90.2 MB 9.0 MB/s eta 0:00:00
Downloading nvidia_cuda_runtime-13.0.96-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (2.2 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.2/2.2 MB 9.7 MB/s eta 0:00:00
Downloading nvidia_cufft-12.0.0.61-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (214.1 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 214.1/214.1 MB 7.4 MB/s eta 0:00:00
Downloading nvidia_cufile-1.15.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (1.2 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.2/1.2 MB 11.0 MB/s eta 0:00:00
Downloading nvidia_curand-10.4.0.35-py3-none-manylinux_2_27_x86_64.whl (59.5 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 59.5/59.5 MB 9.9 MB/s eta 0:00:00
Downloading nvidia_cusolver-12.0.4.66-py3-none-manylinux_2_27_x86_64.whl (200.9 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 200.9/200.9 MB 8.1 MB/s eta 0:00:00
Downloading nvidia_cusparse-12.6.3.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (145.9 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 145.9/145.9 MB 10.0 MB/s eta 0:00:00
Downloading nvidia_nvtx-13.0.85-py3-none-manylinux1_x86_64.manylinux_2_5_x86_64.whl (148 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 148.0/148.0 kB 5.3 MB/s eta 0:00:00
Downloading torchvision-0.28.0-cp312-cp312-manylinux_2_28_x86_64.whl (7.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.7/7.7 MB 3.8 MB/s eta 0:00:00
Downloading ultralytics-8.4.95-py3-none-any.whl (1.4 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.4/1.4 MB 10.0 MB/s eta 0:00:00
Downloading cuda_bindings-13.3.1-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl (6.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.7/6.7 MB 10.3 MB/s eta 0:00:00
Downloading fsspec-2026.6.0-py3-none-any.whl (203 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 203.9/203.9 kB 6.8 MB/s eta 0:00:00
Downloading matplotlib-3.11.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (10.0 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 10.0/10.0 MB 10.5 MB/s eta 0:00:00
Downloading ml_dtypes-0.5.4-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (5.0 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.0/5.0 MB 10.1 MB/s eta 0:00:00
Downloading networkx-3.6.1-py3-none-any.whl (2.1 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.1/2.1 MB 8.3 MB/s eta 0:00:00
Downloading nvidia_ml_py-13.610.43-py3-none-any.whl (53 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 53.2/53.2 kB 3.7 MB/s eta 0:00:00
Downloading opencv_python-5.0.0.93-cp37-abi3-manylinux_2_28_x86_64.whl (73.8 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 73.8/73.8 MB 7.6 MB/s eta 0:00:00
Downloading polars-1.42.1-py3-none-any.whl (837 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 837.6/837.6 kB 9.5 MB/s eta 0:00:00
Downloading polars_runtime_32-1.42.1-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (57.3 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 57.3/57.3 MB 10.5 MB/s eta 0:00:00
Downloading psutil-7.2.2-cp36-abi3-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl (155 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 155.6/155.6 kB 6.3 MB/s eta 0:00:00
Downloading sympy-1.14.0-py3-none-any.whl (6.3 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.3/6.3 MB 8.9 MB/s eta 0:00:00
Downloading ultralytics_thop-2.0.20-py3-none-any.whl (28 kB)
Downloading flatbuffers-25.12.19-py2.py3-none-any.whl (26 kB)
Downloading contourpy-1.3.3-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (362 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 362.6/362.6 kB 6.0 MB/s eta 0:00:00
Downloading cuda_pathfinder-1.5.6-py3-none-any.whl (52 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 53.0/53.0 kB 4.3 MB/s eta 0:00:00
Downloading cycler-0.12.1-py3-none-any.whl (8.3 kB)
Downloading fonttools-4.63.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (5.0 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.0/5.0 MB 11.8 MB/s eta 0:00:00
Downloading kiwisolver-1.5.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (1.5 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.5/1.5 MB 7.6 MB/s eta 0:00:00
Downloading mpmath-1.3.0-py3-none-any.whl (536 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 536.2/536.2 kB 4.7 MB/s eta 0:00:00
Downloading nvidia_nvjitlink-13.3.33-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (40.7 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 40.7/40.7 MB 7.6 MB/s eta 0:00:00
Downloading pyparsing-3.3.2-py3-none-any.whl (122 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 122.8/122.8 kB 6.0 MB/s eta 0:00:00
Installing collected packages: nvidia-ml-py, nvidia-cusparselt-cu13, mpmath, flatbuffers, cuda-toolkit, triton, sympy, pyparsing, psutil, polars-runtime-32, opencv-python, onnxruntime, nvidia-nvtx, nvidia-nvshmem-cu13, nvidia-nvjitlink, nvidia-nccl-cu13, nvidia-curand, nvidia-cufile, nvidia-cuda-runtime, nvidia-cuda-nvrtc, nvidia-cuda-cupti, networkx, ml_dtypes, kiwisolver, fsspec, fonttools, cycler, cuda-pathfinder, contourpy, polars, onnx, nvidia-cusparse, nvidia-cufft, nvidia-cublas, matplotlib, cuda-bindings, nvidia-cusolver, nvidia-cudnn-cu13, torch, ultralytics-thop, torchvision, ultralytics, ls-ml-toolkit
Successfully installed contourpy-1.3.3 cuda-bindings-13.3.1 cuda-pathfinder-1.5.6 cuda-toolkit-13.0.3.0 cycler-0.12.1 flatbuffers-25.12.19 fonttools-4.63.0 fsspec-2026.6.0 kiwisolver-1.5.0 ls-ml-toolkit-1.0.2 matplotlib-3.11.0 ml_dtypes-0.5.4 mpmath-1.3.0 networkx-3.6.1 nvidia-cublas-13.1.1.3 nvidia-cuda-cupti-13.0.85 nvidia-cuda-nvrtc-13.0.88 nvidia-cuda-runtime-13.0.96 nvidia-cudnn-cu13-9.20.0.48 nvidia-cufft-12.0.0.61 nvidia-cufile-1.15.1.6 nvidia-curand-10.4.0.35 nvidia-cusolver-12.0.4.66 nvidia-cusparse-12.6.3.3 nvidia-cusparselt-cu13-0.8.1 nvidia-ml-py-13.610.43 nvidia-nccl-cu13-2.29.7 nvidia-nvjitlink-13.3.33 nvidia-nvshmem-cu13-3.4.5 nvidia-nvtx-13.0.85 onnx-1.22.0 onnxruntime-1.27.0 opencv-python-5.0.0.93 polars-1.42.1 polars-runtime-32-1.42.1 psutil-7.2.2 pyparsing-3.3.2 sympy-1.14.0 torch-2.13.0 torchvision-0.28.0 triton-3.7.1 ultralytics-8.4.95 ultralytics-thop-2.0.20
(label-studio-env) zero@PC-LJM:~/label-studio-project/label-studio-env$安装完毕后确认下：
bash
(label-studio-env) zero@PC-LJM:~/label-studio-project/label-studio-env$ lsml-train --help
╔══════════════════════════════════════════════════════════════╗
║ LS-ML-Toolkit ║
║ Label Studio → YOLO → ONNX ║
║ v1.0.2 ║
╚══════════════════════════════════════════════════════════════╝
usage: lsml-train [-h] [--dataset-name DATASET_NAME] [--dataset-dir DATASET_DIR]
[--epochs EPOCHS] [--imgsz IMGSZ] [--batch BATCH] [--device DEVICE]
[--output-model OUTPUT_MODEL] [--config CONFIG] [--train-split TRAIN_SPLIT]
[--val-split VAL_SPLIT] [--optimize] [--no-optimize] [--force-download]
json_file
Train YOLO model from Label Studio dataset
positional arguments:
json_file Path to Label Studio JSON file
options:
-h, --help show this help message and exit
--dataset-name DATASET_NAME
Dataset name (default: basename of json file)
--dataset-dir DATASET_DIR
Dataset directory (default: from ls-ml-toolkit.yaml or 'dataset')
--epochs EPOCHS Number of training epochs (default: from ls-ml-toolkit.yaml or 50)
--imgsz IMGSZ Image size for training (default: from ls-ml-toolkit.yaml or 640)
--batch BATCH Batch size for training (default: from ls-ml-toolkit.yaml or 8)
--device DEVICE Device for training (default: from ls-ml-toolkit.yaml or auto)
--output-model OUTPUT_MODEL
Output ONNX model path (default: from ls-ml-toolkit.yaml)
--config CONFIG Path to ls-ml-toolkit.yaml file (default: ls-ml-toolkit.yaml)
--train-split TRAIN_SPLIT
Training data split ratio (default: from ls-ml-toolkit.yaml or 0.8)
--val-split VAL_SPLIT
Validation data split ratio (default: from ls-ml-toolkit.yaml or 0.2)
--optimize Enable model optimization after export
--no-optimize Disable model optimization after export
--force-download Force re-download of existing images
Examples:
# Basic training
lsml-train dataset.json --epochs 50 --batch 8
# Training with custom config
lsml-train dataset.json --config custom.yaml --device mps
# Force re-download images
lsml-train dataset.json --force-download --epochs 100
(label-studio-env) zero@PC-LJM:~/label-studio-project/label-studio-env$能整行查询到数据，安装成功
三、训练
bash
lsml-train project.json --epochs 50 --batch 8 --device auto --train-split 1.0 --val-split 0.0