Added Triton deployment instructions to documentation #1116
Conversation
github-actions
bot
added
the
documentation
|
@narendasan can we have this merged before EoW? This is needed for marketing release. |
tanayvarshney
changed the title
|
Changing reviewer to @peri044 |
| # <xx.xx> is the yy:mm for the publishing tag for NVIDIA's Pytorch | ||
| # container; eg. 22.04 | ||
|
|
||
| docker run -it --gpus all -v /path/to/folder:/resnet50_eg nvcr.io/nvidia/pytorch:<xx.xx>-py3 |
There was a problem hiding this comment.
-v /path/to/folder:/resnet50_eg - is this required here ?
There was a problem hiding this comment.
If so, please elaborate in the README as to which folder we should mount to docker
There was a problem hiding this comment.
path/to/local/folder/to/copy/model:/resnet50_eg is this required here ?
Like torch.hub will automatically download the RN50 model right as per model = torch.hub.load('pytorch/vision:v0.10.0', 'resnet50', pretrained=True).eval().to("cuda") ?
There was a problem hiding this comment.
Also, resnet50_eg it is not being used anywhere in the instructions. We don't need this I think
Description
Added documentation about deploying a model to triton
Fixes # (issue)
Type of change
Please delete options that are not relevant and/or add your own.