NOTE: The SHMEM allocation limit is set to the default of 64MB.  This may be

   insufficient for PyTorch.  NVIDIA recommends the use of the following flags:

   docker run --gpus all --ipc=host --ulimit memlock=-1 --ulimit stack=67108864 ...


INFO:     Started server process [1]

INFO:     Waiting for application startup.

INFO:     Application startup complete.

INFO:     Uvicorn running on [http://0.0.0.0:8000⁠](http://0.0.0.0:8000/) (Press CTRL+C to quit)

models/school/coco_classes.txt models/school/yolov3-10.onnx

license provides : ['CPUExecutionProvider']

Loading OnnxLPDetection...

 LP ONNX : 0.04285 sec

Loading Complete

object provides : ['CPUExecutionProvider']

Loading OnnxObjectDetection...

CAR ONNX : 0.27477 sec

Loading Complete

models/school/coco_classes.txt models/school/yolov3-10.onnx

license provides : ['CPUExecutionProvider']

Loading OnnxLPDetection...

FROM rocm/pytorch:rocm6.4_ubuntu24.04_py3.12_pytorch_release_2.6.0

...

RUN pip install pillow \
                protobuf \
                flatbuffers \
                scikit-learn \
                numpy \
                onnxruntime-rocm -f https://repo.radeon.com/rocm/manylinux/rocm-rel-6.4/ \
                mariadb \
                setuptools "uvicorn[standard]" \
                fastapi \
                PyMySQL \
                opencv-python \
                opencv-contrib-python \
                opencv-contrib-python-headless \
                ollama==0.5.3