Skip to content
This repository was archived by the owner on Aug 15, 2025. It is now read-only.

Conversation

jayenashar
Copy link
Contributor

No description provided.

@soumith soumith merged commit faaaf43 into pytorch:master May 18, 2020
@@ -13,7 +13,7 @@ OR
nvidia-docker run -it --ipc=host --rm -v $(pwd):/remote soumith/manylinux-cuda100:latest bash
```

Then run:
Exit the container, then back in the host, run:

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@jayenashar I know this is an old PR, but I'm trying to follow the instructions and the logic, and don't understand... if I run the nvidia-ducker run -it above, and then "exit the container", what's the point of running the docker run? since it isn't detached, exiting it will simply stop the container... furthermore, it seems that the logic in here actually starts another container anyway, so I'm confused about this

# Image
docker_args+=" ${docker_image}"
##############################################################################
# We start a container and detach it such that we can run
# a series of commands without nuking the container
echo "Starting container for image ${docker_image}"
id=$(nvidia-docker run ${docker_args} /bin/cat)

perhaps the instructions are outdated?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

i can't recall and i've given up on building pytorch packages, but i would guess it's to pre-download the image. i didn't write lines 11-13 so i really don't know. you may also notice the --rm which will remove the stopped container.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants