From 56861f987a18660ce0275c692f9ab9bb4d9a06dd Mon Sep 17 00:00:00 2001 From: Surbhi Sharma <su.sharma@fz-juelich.de> Date: Wed, 19 Jun 2024 22:19:55 +0200 Subject: [PATCH] Update file README.md --- README.md | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index c86e63a..5ebe820 100644 --- a/README.md +++ b/README.md @@ -27,9 +27,26 @@ The Remote Sensing Application trains a multispectral (not only RGB channels but ### Application training with Elastic Horovod + Horovod must be installed with **Gloo** to use Elastic Horovod. + Use Scripts **config.py**, **prepare_data_mixup.py**, and **normalization_array** that are available in /scripts folder -+ Run **elastic_training.py** to start training, training and validation scores and losses visible in the output file, available in /Elastic Horovod folder ++ Run **elastic_training.py** to start training, training and validation; scripts available in /Elastic Horovod folder; scores and losses visible in the output file + **jobscript_elastic.sh** is an example job script to run the application with Elastic Horovod on UNITO +### Instructions to run Remote Sensing App with Hercules on Torino cluster using Spack. +#### Install Hercules with Spack. ++spack install hercules + ++ Use Scripts **config.py**, **prepare_data_mixup.py**, and **normalization_array** that are available in /scripts folder + +#### Edit Hercules' configuration file available in /Ad-hoc Storage Systems/Hercules folder. ++ Change the paths + +#### Run the slurm script available in /Ad-hoc Storage Systems/Hercules folder. +sbatch -N3 hercules_torino-1server-1process-in-1node-attached.conf + + +### Instructions to run Remote Sensing App with Expand on Torino cluster. ++ Use Scripts **config.py**, **prepare_data_mixup.py**, and **normalization_array** that are available in /scripts folder ++ Run jobscript_xpn.sh; available in /Ad-hoc Storage Systems/Expand(XPN) folder + ## Contact info For further informations, contact: + su.sharma@fz-juelich.de -- GitLab