Image building fail during executing make prebuild command.

I tried to execute make prebuild command on /closed/NVIDIA directory. How to solve this problem?

775.6 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv 780.0 Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com 780.1 Collecting apex@ git+https://github.com/nvidia/apex@0da3ffb92ee6fbe5336602f0e3989db1cd16f880 (from -r requirements.x86_64.2.txt (lin e 1)) 780.1 Cloning https://github.com/nvidia/apex (to revision 0da3ffb92ee6fbe5336602f0e3989db1cd16f880) to ./pip-install-177uc4fo/apex_06721 f2a06e34e55be55df8d71638642 780.1 Running command git clone --filter=blob:none --quiet https://github.com/nvidia/apex /tmp/pip-install-177uc4fo/apex_06721f2a06e34e5 5be55df8d71638642 781.6 Running command git rev-parse -q --verify 'sha^0da3ffb92ee6fbe5336602f0e3989db1cd16f880' 781.6 Running command git fetch -q https://github.com/nvidia/apex 0da3ffb92ee6fbe5336602f0e3989db1cd16f880 782.1 Running command git checkout -q 0da3ffb92ee6fbe5336602f0e3989db1cd16f880 783.2 Resolved https://github.com/nvidia/apex to commit 0da3ffb92ee6fbe5336602f0e3989db1cd16f880 783.2 Running command git submodule update --init --recursive -q 793.7 Preparing metadata (setup.py): started 795.3 Preparing metadata (setup.py): finished with status 'error' 795.3 error: subprocess-exited-with-error 795.3 795.3 × python setup.py egg_info did not run successfully. 795.3 │ exit code: 1 795.3 ╰─> [8 lines of output] 795.3 Traceback (most recent call last): 795.3 File "", line 2, in 795.3 File "", line 34, in 795.3 File "/tmp/pip-install-177uc4fo/apex_06721f2a06e34e55be55df8d71638642/setup.py", line 10, in 795.3 from torch.utils.cpp_extension import BuildExtension, CppExtension, CUDAExtension, CUDA_HOME, load 795.3 File "/usr/local/lib/python3.8/dist-packages/torch/utils/cpp_extension.py", line 25, in 795.3 from pkg_resources import packaging # type: ignore[attr-defined] 795.3 ImportError: cannot import name 'packaging' from 'pkg_resources' (/usr/local/lib/python3.8/dist-packages/pkg_resources/_init _.py) 795.3 [end of output] 795.3 795.3 note: This error originates from a subprocess, and is likely not a problem with pip. 795.3 error: metadata-generation-failed 795.3 795.3 × Encountered error while generating package metadata. 795.3 ╰─> See above for output. 795.3 795.3 note: This is an issue with the package mentioned above, not pip. 795.3 hint: See above for details.

Dockerfile.x86_64:129

128 | # Break requirements into two lists because some of them require that other packages be fully installed first. 129 | >>> RUN python3 -m pip install -r requirements.x86_64.1.txt \ 130 | >>> && python3 -m pip install -r requirements.x86_64.2.txt 131 |

ERROR: failed to solve: process "/bin/bash -c python3 -m pip install -r requirements.x86_64.1.txt && python3 -m pip install -r requiremen ts.x86_64.2.txt" did not complete successfully: exit code: 1 make[1]: [Makefile.docker:259: build_docker] Error 1 make[1]: Leaving directory '/mnt/sdb/mlperf/code/inference_results_v3.1/closed/NVIDIA' make: [/mnt/sdb/mlperf/code/inference_results_v3.1/closed/NVIDIA/Makefile.docker:247: prebuild] Error 2

mlcommons / inference_results_v3.1

Image building fail during executing make prebuild command. #21

Dockerfile.x86_64:129

128 | # Break requirements into two lists because some of them require that other packages be fully installed first. 129 | >>> RUN python3 -m pip install -r requirements.x86_64.1.txt \ 130 | >>> && python3 -m pip install -r requirements.x86_64.2.txt 131 |