This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
If you have a model that takes longer than 10 seconds to return results, by default Promote will time out your model API query. If you need Promote to wait longer than 10 seconds before timing out, you can adjust this time out setting with an environment variable called PREDICTION_TIMEOUT. A model will wait a maximum of 60 seconds before timing out.
Adding an environment variable to a model is easy! First, open the model for which you would like to modify the timeout variable by clicking the model's name.
After opening the dashboard, navigate to the Advanced tab, and find the row for environment variables. Click on the small plus sign in this row to add an environment variable.
In the modal window that pops up, add PREDICTION_TIMEOUT as the variable Name. Add the length of time Promote should wait before timing out a prediction (in seconds) as the Value, then click Save. Although the Environment Variable will accept any number you provide as a Value, the maximum time a model will wait before timing out is 60 seconds.
You should see your new environment variable populated in the Environment Variables row of the Advanced tab for your model.
If you need to revert this setting to the default or modify the value, you will first need to delete the environment variable you created (you can not edit an environmentvariable once it has been saved). To delete an environment variable, click on the little cog icon next to the variable in the dashboard view, and select delete.
Once the variable is deleted, you can create another PREDICTION_TIMEOUT variable with a new value.
Other Environment Variables
There is currently one other built-in environment variable, specifically for R Promote models, that allows you to toggle between JSON packages. To read more about it, please see this Community Knowledge Basearticle.