ai-center
2021.10
false
- Getting Started
- Network requirements
- Single-node requirements and installation
- Multi-node requirements and installation
- Post-installation
- Accessing AI Center
- Provision an AI Center tenant
- Updating Orchestrator and Identity Server certificates
- Resizing PVC
- Adding a new node to the cluster
- ML packages offline installation
- Configuring the Cluster
- Configuring the FQDN post-installation
- Backing up and Restoring the Cluster
- Using the Monitoring Stack
- Setting up a Kerberos Authentication
- Provisioning a GPU
- Using the configuration file
- Node scheduling
- Migration and Upgrade
- Basic Troubleshooting Guide
ML packages offline installation
OUT OF SUPPORT
AI Center Installation Guide
Last updated Nov 11, 2024
ML packages offline installation
Note: In order to use ML Packages, you must first upload the packages in AI Center. To do so, follow the steps from Upload model to AI Center. Some models, like Document Understanding, need to be installed. If you plan on using these models, make sure to install
them first by following the procedure from Install the offline bundle.
For Document Understanding ML Packages offline installation, please check this page.
To install out of the box packages in an airgapped environment, follow the steps below.
- Access Out of the Box Packages at tenant level. To add a new ML Package, click the Create new button.
Note: In order to access the page, make sure that you have Read permissions. The Create new button is only visible if you have Create permissions as well.
- Go to the ML Logs page. The ML packages uploaded and validated from the public tenant are displayed here. Click Create New to upload a new model.
- Upload the ML package
zip
file andjson
file separately. - Click Submit.
Note: After you upload the model, wait a few minutes until validation is successful. After validation is successful, refresh the grid manually. You can now go to any project in the Out of the box packages page and you can find the newly uploaded ML Package.
Once the ML package is uploaded, the ML Logs page is displayed.
- Before installing the bundle make sure it is not already there. If that's not the case or if you are unsure you can proceed to next step, otherwise you just need to follow upload model documentation.
- For Windows machines, directly download through the bundle link. For Linux machines, from the machine having access to the
internet, download the needed bundle following the below command:
wget -O ~/<bundle-name.tar.gz> 'bundle-link'
wget -O ~/<bundle-name.tar.gz> 'bundle-link' - Copy the following bundle to the
/uipath/tmp
folder on the main machine of the cluster (where the install took place):scp ~/<bundle-name.tar.gz> <username>@<node dns>:/uipath/tmp/
scp ~/<bundle-name.tar.gz> <username>@<node dns>:/uipath/tmp/ - Connect to this main machine and load the bundle:
./install-uipath.sh -i ./cluster_config.json -o ./output.json --optional-offline-bundle "/uipath/tmp/<bundle-name.tar.gz>" --offline-tmp-folder "/uipath/tmp" --accept-license-agreement
./install-uipath.sh -i ./cluster_config.json -o ./output.json --optional-offline-bundle "/uipath/tmp/<bundle-name.tar.gz>" --offline-tmp-folder "/uipath/tmp" --accept-license-agreement