我尝试使用 API 文档中描述的以下命令在 Azure 机器学习上安装带有 Tensorflow 2 的对象检测 API :
git clone https://github.com/tensorflow/models.git
docker build -f research/object_detection/dockerfiles/tf2/Dockerfile -t od .
docker run -it od
cd models/research
protoc object_detection/protos/*.proto --python_out=.
cp object_detection/packages/tf2/setup.py .
python -m pip install --use-feature=2020-resolver .
但是当我执行上面(python -m pip install ...
)的最后一个代码时,我得到了以下错误。错误太长,错误的每一行都说[Errno 38] Function not implemented:
. 错误的尾部如下所示:
....(omitted)...
('/mnt/batch/tasks/shared/LS_root/mounts/clusters/(some_path)/models/research/vid2depth/WORKSPACE',
'/tmp/pip-req-build-uxkxrn35/vid2depth/WORKSPACE',
"[Errno 38] Function not implemented: '/mnt/batch/tasks/shared/LS_root/mounts/clusters/(some_path)/models/research/vid2depth/WORKSPACE'"),
('/mnt/batch/tasks/shared/LS_root/mounts/clusters/(some_path)/models/research/vid2depth',
'/tmp/pip-req-build-uxkxrn35/vid2depth',
"[Errno 38] Function not implemented: '/mnt/batch/tasks/shared/LS_root/mounts/clusters/(some_path)/models/research/vid2depth'"),
('/mnt/batch/tasks/shared/LS_root/mounts/clusters/(some_path)/models/research',
'/tmp/pip-req-build-uxkxrn35',
"[Errno 38] Function not implemented: '/mnt/batch/tasks/shared/LS_root/mounts/clusters/(some_path)/models/research'")]
我已经确认在我的本地 PC 上安装了相同的命令是成功的。如果我想在 Azure 机器学习上安装它,我需要做一些不同的事情吗?