Moving from pynvml to nvidia-ml-py

[jakirkham]

Currently this is making use of pynvml in a few places:

cloud-ml-examples/gcp/docker/infrastructure/rapids_lib.py

Line 29 in 5692c16

import pynvml

cloud-ml-examples/gcp/docker/infrastructure/rapids_lib.py

Lines 248 to 251 in 5692c16

	pynvml.nvmlInit()
	for iGPU in range(self.n_workers):
	handle = pynvml.nvmlDeviceGetHandleByIndex(iGPU)
	print_device_memory(pynvml.nvmlDeviceGetMemoryInfo(handle))

https://github.com/rapidsai/cloud-ml-examples/blame/5692c1655cd2581d696fdba0fb503d18097cddff/gcp/notebooks/custom_hpo.ipynb#L81

cloud-ml-examples/azure/code/rapids_csp_azure.py

Line 38 in 5692c16

import pynvml

However we would like to move to nvidia-ml-py in the future. Raising this issue to track this work

[jacobtomlinson]

Thanks John! This repo is in the process of being archived and content being migrated to rapidsai/deployment so I'm going to close this in favour of rapidsai/deployment#151

[jakirkham]

Ah ok. Sorry for the noise then

There was a bit of gray area when looking through these repos to ascertain, which still needed attention