- 10 Sep 2024
- 3 Minutes to read
- Print
- DarkLight
- PDF
Use LandingEdge with LandingLens on Snowflake
- Updated on 10 Sep 2024
- 3 Minutes to read
- Print
- DarkLight
- PDF
This article applies to these versions of LandingLens:
LandingLens | LandingLens on Snowflake |
✖ | ✓ |
Users of both LandingLens and LandingLens on Snowflake can use LandingEdge to run inference.
Most of the LandingEdge processes are the same for both LandingLens and LandingLens on Snowflake—you choose an image source, an action that triggers inference, and the LandingLens model that runs inference.
The only difference occurs if you want to run LandingEdge "online". This means that LandingEdge uses an Internet connection to load the model from LandingLens on Snowflake. Running LandingEdge online also allows you to upload inference results back to LandingLens on Snowflake, so that you can save images to your project in a process called Continuous Learning.
If you want to run LandingEdge online with LandingLens on Snowflake, follow the instructions in this article. Otherwise, if you want to run LandingEdge "offline", follow the general instructions in Manage Inspection Points.
Considerations for Running LandingEdge Online with LandingLens on Snowflake
If you enable running LandingEdge online with LandingLens on Snowflake, you will only be able to use LandingEdge to run models online with LandingLens on Snowflake. You won't be able to run models offline or run models from the LandingAI-hosted version of LandingLens until you disable the connection with LandingLens on Snowflake.
Set Up an Online Connection with LandingLens on Snowflake
If you want to run LandingEdge online with LandingLens on Snowflake, set up a connection between the two applications using the following instructions:
- Open LandingEdge.
- Click the Settings (gear) icon in the top right corner of the LandingEdge homepage.
- Click Snowflake.
- Select the Use Snowflake checkbox.
- Fill out the fields:
- Snowflake App URL: Enter the fully qualified domain of your instance of LandingLens on Snowflake. For example, the URL could be
https://example.snowflakecomputing.app/
. - Snowflake Account: Enter your Snowflake account locator. One way to find your account locator is to open a worksheet in Snowsight and run
SELECT CURRENT_ACCOUNT();
. For more information, go to the Snowflake documentation. - Snowflake Username: Enter your Snowflake username as it appears in your Snowsight profile. (This is the
NAME
property in Snowflake.)
- Snowflake App URL: Enter the fully qualified domain of your instance of LandingLens on Snowflake. For example, the URL could be
- Click Generate New Private Key. LandingEdge creates a private key and saves it to your hard drive.
- Click Copy Public Key. (You will later assign the public key to yourself in Snowflake.)
- If you're using AWS PrivateLink or your Snowflake account locator has underscores, select the Use AWS PrivateLink checkbox.
- Click Save.
- Assign the public key to yourself in Snowflake. To learn how to do this, go to the Snowflake documentation.
- To verify that LandingEdge successfully connected to LandingLens on Snowflake, set up an inspection point. If your projects display in the Project Connected to drop-down menu, the connection works. If you don't see any projects, check if all Snowflake account information was entered correctly.
Set Up an "Online" Inspection Point
After you've set up the Snowflake settings in LandingEdge, you can set up Inspection Points, which are the sets of information needed to deploy models in LandingEdge. For the most part, you can follow the instructions in Manage Inspection Points to set up an Inspection Point. The only difference is how you fill out the Cloud Connection and Model settings:
- Select your project from the Project Connected to drop-down menu.
- Select Download from LandingLens.
- Click Select.
- Select the model you want to use and click Download Model.
- LandingEdge loads the model. When this is complete, the Model successfully loaded message displays and the Run Now button becomes active.
- Continue setting up the Inspection Point.
Disable the Online Connection with LandingLens on Snowflake
If you no longer want LandingEdge to be connected to LandingLens on Snowflake, deselect the Use Snowflake checkbox on LandingEdge's Snowflake settings page. This disconnects all Inspection Points that were connected to LandingLens on Snowflake.