install failed for 22.09 (works okay for 22.08)

[jdunham22]

System information

• OS Platform and Distribution: Docker host: Ubuntu 18.04.1, docker-ce 5:20.10.2
• TensorFlow installed from (source or binary): binary (pip)
• TensorFlow version: 1.15.5+nv22.09
• Python version: 3.8.15
• Installed using virtualenv? pip? conda?: pip
• Bazel version (if compiling from source): n/a
• GCC/Compiler version (if compiling from source): n/a
• CUDA/cuDNN version: none
• GPU model and memory: none

problem: current version 1.15.5+nv22.09 does not install We are using tensorflow in inference mode on a system without a gpu, though I don't know if this error is caused by the lack of the GPU and drivers.

command: docker run --tty --rm python:3.8.15 sh -c "pip install nvidia-pyindex && pip install nvidia-tensorflow==1.15.5+nv22.09"

output: ERROR: Could not find a version that satisfies the requirement nvidia-nccl-cu11~=2.15 (from nvidia-tensorflow) (from versions: 0.0.1.dev5, 2.13.4, 2.14.3, 2021.10.26, 2021.11.16, 2022.3.30, 2022.3.31, 2022.5.19) ERROR: No matching distribution found for nvidia-nccl-cu11~=2.15

[bertsky]

Same happens on a system (also Python 3.8.15, but without Docker) with GPU and CUDA 11.7, so that's not the problem.

My guess is: the dependency nvidia-nccl-cu11~=2.15 must be replaced with a version string following the new schema.