Skip to content

Setup

Please replace the <placeholder> with the corresponding value:

  • <workspace_name>: Name of the folder in /opt/hpe/swarm-learning-hpe/workspace where the configuration files and model code is stored. e.g.: odelia-breast-mri
  • <sentinel_ip>: IP address of the sentinel host. The sentinel host is the initiator of the network and the operator of the license server in our case. e.g.: 172.24.4.67
  • <host_index>: Name of the institution or partner participating in the training e.g.: TUD

Installlation of HPE Swarm Learning

Please only proceed to the next step by observing "... is done successfully" from the log.

/opt/hpe/swarm-learning-hpe

Please make sure you are in /opt/hpe/swarm-learning-hpe before you start the installation. You can get there with cd /opt/hpe/swarm-learning-hpe.

1. Prerequisite:

Runs scripts that check for required software and opens exposed ports.

sh workspace/automate_scripts/automate.sh -a

2. Server setup:

Runs scripts that sets up the swarm learning environment on a server.

sh workspace/automate_scripts/automate.sh -b -s <sentinel_ip> -d <host_index>

Credentials

  • The script will ask for the HPE credentials. Please use your HPE Account credentials.
  • The script will also ask for the VPN credentials. We use GoodAccess VPN. ODELIA consortium members get their credentials through the contact persons at TU Dresden. External users are welcome to open an issue on Github if they have any questions.

3. Final setup:

Runs scripts that finalize the setup of the swarm learning environment.

sh workspace/automate_scripts/automate.sh -c -w <workspace_name> -s <sentinel_ip> -d <host_index> -l <license_ip> [-n num_peers] [-e num_epochs]

Optional flags

  • -n num_peers: Number of peers to be added to the network. e.g.: 3
  • -e num_epochs: Number of epochs to be trained. e.g.: 10

Upgrade the Swarm Learning Environment

  1. If you already have the old version of Swarm learning installed. Run the following command to upgrade the Swarm Learning Environment from 1.x.x to 2.x.x

    sh workspace/automate_scripts/server_setup/cleanup_old_sl.sh
    

  2. Then proceed to 1. Prerequisite in the installation guide.

Bugs and Problems

Did you find a bug in the code or other problems? Then raise an issue in our Github repository: https://github.com/KatherLab/swarm-learning-hpe/issues

In case of problems or requests for improvement of the documentation, please raise an issue at: https://github.com/odelia-ai/odelia-ai.github.io/issues