Perform the following steps to deploy an interactive Streamlit dashboard for classifying (labeling) and browsing collected images in your Iguazio Data Science Platform ("the platform") environment:
-
On you platform, launch a Jupyter Notebook service, navigate to "demos/realtime-face-recognition/streamlit" and locate the file streamlit.yaml
-
right click on streamlit.yaml and select Download
-
Ensure that you have a trained model, generated by using the face-recognition.ipynb demo notebook.
-
Run the following command from a command-line shell on your local computer to deploy the Streamlit dashboard as a service; ensure that the current directory has your streamlit.yaml file that you have downloaded or change the path (
-f) to point to your file:kubectl -n default-tenant apply -f streamlit.yaml
Note: To successfully run this
kubectlcommand you need a kubeconfig file with the required permissions. Contact Iguazio support for assistance. -
In your web browser, go to port 30090 of your platform's application node —
<your platform application-node IP address>:30090— and view your interactive Streamlit dashboard:
