Google Sheets
Deploying to Google Sheets
Last updated
Deploying to Google Sheets
Last updated
To deploy your model to google sheets, you select the Google Sheet option in the deployment screen. The example shown here takes data from Google Sheets and Deploys back to the same instance, but you can have your source and deployments in different systems.
Once selected, there are several options for a successful deployment.
Select the sheet you will be putting your data to predict against. This sheet needs to have already been created. You may need to refresh the window if you created it since opening the Akkio project. Using the same two-step process as in getting your initial data, you will want to select connect dataset to navigate to the spreadsheet you are using and then click it again once it has loaded into Akkio.
Important!
Your sheet must have some additional columns for Akkio to place predictions. Akkio will not widen your sheet for you.
Scheduler - Select how often the prediction will run
Run on Deploy - Select whether it will run immediately upon deployment.
Map Fields - Map the fields existing in the flow to the fields on the prediction data. These will often be the same but can be manually mapped if they are not identical (for example, here, Job Title could have been simply 'Job' in the Google Sheet).
Apply Data Prep - If you applied any data prep steps in the flow, those transformations can be applied to your prediction data before running the prediction.
Upper Prediction Threshold - Applied prediction threshold to the data before outputting predictions. This means you can set a max value for the prediction to be able to return. For numerical predictions only.
Lower Prediction Threshold - Applied prediction threshold to the data before outputting predictions. This means you can set a min value for the prediction to be able to return. For numerical predictions only.
Once you are happy with your settings, you can click to preview and then deploy your model. Once deployed, your Google sheet will update immediately or at the selected interval.
See the sample output for this flow here.
