How to add new data and improve (re-train) your model?

So once you have your model on Hornbill.AI, you can either choose to keep running with that and access it by our various endpoints (AIs, Slack, Discord, etc.) or you can improve it by adding new data.

In this tutorial, we illustrate a 2-step process to improve your model by adding new data. If you haven’t made a model yet, go to our tutorials section and proceed as per the tutorial which best fits your use case.

Step 1

Interact with the datasets that you want to add the new data to. Here we have made added new stock prices to our MSFT_Stock_Price dataset. Make sure that there is no change in the column structure and both the number of columns as well as the datatype is the same as the deployed dataset.

Click on the Add new data tab.

Choose the dataset which you want to upload and then press open. In a couple of seconds, your chosen dataset is added to the deployed dataset.

Step 2

Go to the Projects and select the project for which you added new data. Here that project is MSFT Stock Price Prediction. Just click on Re-train Project.

Hornbill.AI will retrain the project and replace the model internally. The new model will factor in the new data.

Go build on Hornbill.AI, it’s easier than Excel sheets!