release 1.12.1

Release 1.12.1: Enhanced User Journey and New Data Connectors

December 2023 marked the release of 1.12.1: Enhanced data connections, a revamped Customer churn template, and user-friendly model building.


Hello data enthusiasts! We're excited to share the scoop on our newest release, the AI & Analytics Engine version 1.12.1. Loaded with user-centric features, this update is geared towards ironing out those user journey wrinkles and delivering a more seamless experience overall. So, without further ado, let's delve into the specifics and see how these enhancements are set to elevate your analytics journey!

1. Customer Churn Template: Predict with Precision 📊

Say goodbye to validation errors! We've revamped the Customer churn template to offer a more logical and streamlined experience. Now, in the very first step, you specify your business requirements, and we guide you on the optimal time span of data needed. No more post-import surprises, you can now predict with confidence.

Customer Churn Template

2. New Import and Export Options for Seamless Data Integration ⛓️

In response to your feedback, we've expanded our data connectivity options. This means that you can now connect to your datasets to the AI & Analytics Engine from an extensive list of sources.

New import options:

  • Google Sheets

  • AWS Redshift

  • Google BigQuery

New export options:

  • AWS RDS

  • AWS Redshift

  • Google Cloud SQL

  • Google BigQuery

 New Import and Export Options for the AIAE

3. Build from Scratch: Streamlining Your Model by Excluding Irrelevant Features 🛠️

Building your model from scratch has never been more flexible. Introducing new options in the third step of the Build from scratch app builder, allowing you to exclude irrelevant features. Streamline your analysis, training, and prediction by excluding unwanted features such as irrelevant data and target leakage columns.

Excluding unwanted features in automatic feature selectionExcluding unwanted features in automatic feature selection

Excluding unwanted features before estimating feature importance scores to aid manual feature selectionExcluding unwanted features before estimating feature importance scores to aid manual feature selection

 

4. Training New Models Within an App: A User-Friendly Approach 🚀

Embark on a smooth user experience with the new Train new model function. You can now:

  • Re-use a previous feature selection or manually choose a new set of features.

  • Specify your requirements in terms of prediction quality and training time, and witness immediate recommendations from the model recommender.

  • The introduction of "training requests" keeps you informed about background tasks, providing transparency into the model-building process.

 

With the release of AI & Analytics Engine 1.12.1, we've not improved the flow for our users but also focused on elevating your experience. Unleash the potential of seamless insights, tailored analytics, and improved data connections. Stay tuned for more exciting updates in the future! 🌟

Similar posts

Subscribe to PI.EXHANGE emails!

Get the latest news, articles, and thought pieces direct to your inbox