Basic repository showing how to use Hydra + Hydra launchers on SLURM cluster

Overview

Slurm-Hydra-Submitit

This repository is a minimal working example on how to:

Set up Hydra

⚠️ You need to install hydra-core for this step.

Hydra is fairly easy to set-up:

By simply running python slurm_hydra_submitit/script.py, you'll see how the main function takes the arguments from the configuration file and pass them to the following underlying functions.

Launch jobs on a SLURM cluster with Hydra submitit launcher

Launch a job on the cluster

⚠️ You need to install hydra-submitit-launcher for this step.

Now that our Hydra conf is setup, we want to run the job on a SLURM cluster instead of our local computer. For that, we need to:

  • specify the hydra launcher to work on the SLURM cluster
  • specify the hardware specifications for the SLURM job

If you connect to your SLURM cluster scheduler node, just by installing hydra-submitit-launcher, you can already launch jobs on the cluster with:

python slurm_hydra_submitit/script.py --multirun hydra/launcher=submitit_slurm

To test locally before sending to the cluster, you can switch the hydra/launcher argument to submitit_local.

Adapt node parameters

You can easily adapt the SLURM parameters by modifying the following arguments SLURM launcher arguments.

For example, the following script is executed on nodes with 10 CPUs: python slurm_hydra_submitit/script.py --multirun hydra/launcher=submitit_slurm hydra.launcher.cpus_per_task=10

Launch array of jobs on the SLURM cluster

Grid Search

You can launch multiple jobs at once by specifying their values in the launch command.

For example, the following command launches 4 jobs which corresponds to all the possible combinations of arguments.

python slurm_hydra_submitit/script.py --multirun hydra/launcher=submitit_slurm project_name=P1,P2 train.epochs=30,40

Specific Parameters Combinations

Alternatively, you can pass sets of parameters to test together:

python slurm_hydra_submitit/script.py --multirun hydra/launcher=submitit_slurm +compile="{project_name:P1,train.epochs:30}, {project_name:P2,train.epochs:40}"

To clean this command a bit, we can create a bash script similar to this:

#!/bin/bash
params=(
    '{project_name:P1,train.epochs:10},'
    '{project_name:P2,train.epochs:20}'
)

slurm_hydra_submitit/script.py --multirun hydra/launcher=submitit_slurm +compile="${params[*]}"
Owner
Raphael Meudec
PhD candidate @Parietal-INRIA
Raphael Meudec
An extension for Arma 3 that lets you write extensions in Python 3

An Arma 3 extension that lets you to write python extensions for Arma 3. And it's really simple and straightforward to use!

Lukasz Taczuk 48 Dec 18, 2022
CPython extension implementing Shared Transactional Memory with native-looking interface

CPython extension implementing Shared Transactional Memory with native-looking interface

21 Jul 22, 2022
Contains the code of my learning of Python OOP.

OOP Python This repository contains the code of my learning of Python OOP. All the code: is following PEP 8 ✅ has proper concept illustrations and com

Samyak Jain 2 Jan 15, 2022
A pypi package details search python module

A pypi package details search python module

Fayas Noushad 5 Nov 30, 2021
A feed generator. Currently supports generating RSS feeds from Google, Bing, and Yahoo news.

A feed generator. Currently supports generating RSS feeds from Google, Bing, and Yahoo news.

Josh Cardenzana 0 Dec 13, 2021
Woltcheck - Python script to check if a wolt restaurant is ready to deliver to your location

woltcheck Python script to check if a wolt restaurant is ready to deliver to you

30 Sep 13, 2022
Semantic Data Management - Property Graphs 📈

SDM - Lab 1 @ UPC 👨🏻‍💻 Table of contents Introduction Property Graph Dataset 1. Introduction This repo is all about what we have done in SDM lab 1

Mohammad Zain Abbas 1 Mar 20, 2022
Ronin - Create Fud Meterpreter Payload To Hack Windows 11

Ronin - Create Fud Meterpreter Payload To Hack Windows 11

Dj4w3d H4mm4di 6 May 09, 2022
Tutor plugin for integration of Open edX with a Richie course catalog

Richie plugin for Tutor This is a plugin to integrate Richie, the learning portal CMS, with Open edX. The integration takes the form of a Tutor plugin

Overhang.IO 2 Sep 08, 2022
Exactly what it sounds like, which is something rad

EyeWitnessTheFitness External recon got ya down? That scan prevention system preventing you from enumerating web pages? Well look no further, I have t

Ellis Springe 18 Dec 31, 2022
laTEX is awesome but we are lazy -> groff with markdown syntax and inline code execution

pyGroff A wrapper for groff using python to have a nicer syntax for groff documents DOCUMENTATION Very similar to markdown. So if you know what that i

Subhaditya Mukherjee 27 Jul 23, 2022
An end-to-end Python-based Infrastructure as Code framework for network automation and orchestration.

Nectl An end-to-end Python-based Infrastructure as Code framework for network automation and orchestration. Features Data modelling and validation. Da

Adam Kirchberger 15 Oct 14, 2022
Pytorch implementation of "Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Rates"

Peer Loss functions This repository is the (Multi-Class & Deep Learning) Pytorch implementation of "Peer Loss Functions: Learning from Noisy Labels wi

Kushal Shingote 1 Feb 08, 2022
An optional component handler for hikari, inspired by discord.py's views.

hikari-miru An optional component handler for hikari, inspired by discord.py's views.

43 Dec 26, 2022
AMTIO aka All My Tools in One

AMTIO AMTIO aka All My Tools In One. I plan to put a bunch of my tools in this one repo since im too lazy to make one big tool. Installation git clone

osintcat 3 Jul 29, 2021
Blender 2.80+ Timelapse Capture Tool Addon

SimpleTimelapser Blender 2.80+ Timelapse Capture Tool Addon Developed for Blender 3.0.0, tested working on 2.80.0 It's no ZBrush undo history but it's

4 Jan 19, 2022
⚙️ Compile, Read and update your .conf file in python

⚙️ Compile, Read and update your .conf file in python

Reece Harris 2 Aug 15, 2022
Python module used to generate random facts

Randfacts is a python library that generates random facts. You can use randfacts.get_fact() to return a random fun fact. Disclaimer: Facts are not gua

Tabulate 14 Dec 14, 2022
Tool to generate wrappers for Linux libraries allowing for dlopen()ing them without writing any boilerplate

Dynload wrapper This program will generate a wrapper to make it easy to dlopen() shared objects on Linux without writing a ton of boilerplate code. Th

Hein-Pieter van Braam 25 Oct 24, 2022
This is a simple leaderboard for 30 days of Google Cloud program for students of ASIET

30daysleaderboard #Hacktoberfest - Please don't make changes in readme file. Only improvement in the project will be accepted. Update - Now if you run

5 Oct 29, 2021