How to further improve Sagemaker Canvas Model Score?

Hi Janice,

Given that Canvas is a no-code tool that abstracts and automate most of the model training process. Model tuning and training performance are mostly limited to the amount of data + your features selection when you train the model.

Nonetheless, one key option you can adopt in improving your score is at the pre-processing stage (i.e feature engineering). Assuming you aren't a technical developer, you can explore the use of AWS Data Brew.

AWS Glue DataBrew is a no-code visual data preparation tool that makes it easy for data analysts and data scientists to clean and normalize data to prepare it for analytics and machine learning. You can choose from over 250 pre-built transformations to automate data preparation tasks, all without the need to write any code. You can automate filtering anomalies, converting data to standard formats, and correcting invalid values, and other tasks. After your data is ready, you can immediately use it for analytics and machine learning projects.

In your case, you can store your initial data in S3 and have databrew performs feature engineering on it before writing it back to S3 again. From there, you can import the processed data back into Canvas and build your model from there. This should gives you a better score compared to building your model directly using the initial raw data.

Data Brew Getting Started guide : Link

Cheers.

Vincent