pytorch/serve

Replace `install_dependencies.py` by `pip install torchserve[cu111]`

Open

#1,516 opened on Mar 16, 2022

View on GitHub
 (5 comments) (0 reactions) (0 assignees)Java (3,844 stars) (790 forks)batch import
documentationhelp wanted

Description

Hi, I'm using the package on a computer with a GPU, and the server fails when trying to import nvgpu. Any reason why this package is not in the requirements ? At least in extras ?

Thanks

Contributor guide

Tech stack
pythonpytorch
Domain
backendmachine learningapi
Issue type
bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
under 1 hour
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
None
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
85
Research direction
The issue reports that the import of nvgpu in `ts/metrics/system metrics.py` line 61 fails because nvgpu is not listed in `setup.py` line 160 extras or requirements. The fix involves adding nvgpu to the extras require in `setup.py` (e.g., as part of a new extra like 'gpu') or updating the installation instructions to use `pip install torchserve[cu111]` as the title suggests. The comments (5) may contain further guidance from maintainers. The contributor should check the linked files and ensure compatibility.