https://github.com/ray-project/scalable-learning
Scaling multi-node multi-GPU workloads
https://github.com/ray-project/scalable-learning
Last synced: about 1 year ago
JSON representation
Scaling multi-node multi-GPU workloads
- Host: GitHub
- URL: https://github.com/ray-project/scalable-learning
- Owner: ray-project
- License: apache-2.0
- Created: 2017-09-12T23:04:02.000Z (almost 9 years ago)
- Default Branch: master
- Last Pushed: 2017-09-13T00:24:39.000Z (almost 9 years ago)
- Last Synced: 2025-03-05T08:45:51.344Z (over 1 year ago)
- Size: 5.86 KB
- Stars: 5
- Watchers: 5
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# scalable-learning
Scaling multi-node multi-GPU workloads
## Installing Imagenet
The commands below assume Python is Python 2, so you may need to change the python invoked in the script if your default Python is Python 3
# Assuming we are starting in repository root
ROOT="$PWD"
cd /tmp
git clone https://github.com/SaMnCo/dl-training-datasets
cd imagenet/ilsvrc14
./00-build.sh "$ROOT/data"
TODO: make own script for this, without the need for the user to specify Python 2.
## Running pytorch example
[https://github.com/pytorch/examples/tree/master/imagenet](Follow steps here).
TODO: copy in relevant files
## Dependencies
TODO: make user-independent aws-magic analogue, pull in docker for pytorch.