⚡Runs
Run (and sweep) any private or public Github repository.

Runs and sweeps

Run (and sweep) any private or public Github repository on the cloud.
1
# clone repo
2
git clone https://github.com/williamFalcon/hello
3
4
# start the sweep
5
cd hello
6
grid run hello.py --number "[1, 2]" --food_item "['pizza', 'hotdog']"
Copied!
⚡️⚡️Forget about infrastructure ⚡️⚡️
Runs are "serverless" which means you only pay for the time your scripts are actually running. When running on your own infrastructure this amounts to massive cost savings as well.

1 minute overview

In this video we're going to run an arbitrary model (from the pytorch examples github repo) across 4 GPUs (4 experiments each on 2 GPUs)

Product Tour

Option 1: Run via the CLI

RUN any GitHub file with Grid in 4 steps:
1
# 1. clone the repo
2
git clone https://github.com/pytorch/examples
3
4
# 2. find the file to run
5
cd examples/dcgan
6
7
# 3. verify it works locally (optional)
8
python main.py --dataset cifar10 --lr 0.0002 --dataroot .
9
10
# 4. run on a cloud instance via grid
11
grid run main.py --dataset cifar10 --lr 0.0002 --dataroot .
Copied!
Grid offers advanced syntax for starting a run. With this code:
1
grid run hello.py --number "[1, 2]" --food_item "['pizza', 'hotdog']"
Copied!
Grid will run the script 4 times... these are the 4 equivalent script calls (we call each script call an experiment)
1
python hello.py --number 1 --food_item 'pizza'
2
python hello.py --number 2 --food_item 'pizza'
3
4
python hello.py --number 1 --food_item 'hotdog'
5
python hello.py --number 2 --food_item 'hotdog'
Copied!
A RUN is a collection of experiments (the run has 4 experiments in this example).

Option 2: Start via the web UI

Last modified 4mo ago