We are excited to announce the latest release of the AI & Analytics Engine - 1.6.0. This release brings about new features, like the Model Leaderboard, Performance Report, and more! The objective of this release is to better the model training and evaluation process.
The intention of the Model Leaderboard is to give users a hand in selecting the most appropriate models for them to deploy after they are successfully trained. In the Model Leaderboard, users can view and compare the models in one of the following criteria:
Best prediction quality - The model with the best performance.
Shortest prediction time - The fastest model to generate predictions for 1000 rows of data
Shortest training time - The model with the shortest training time
This allows users to identify the most appropriate model for their own needs to deploy.
The performance report is a translation of the Engine's findings into comprehensible and precise messages. Users can add, edit, and delete breakpoints in the Prediction Error Breakdown table to see the model’s performance on specific value ranges.
Our brand new What-If analysis feature helps users:
Enter individual samples to make prediction outputs;
Compare these individual inputs and outputs side by side; and
Use their domain expertise to evaluate how well their model performed and how to best use their models.
The Engine applies state-of-the-art algorithms to automatically suggest a list of the most impactful columns for your classification and regression models. Users save time by skipping the manual selection process, with the recommended feature set automatically generated.
Our engineers have worked around the clock to improve the model performance. With Model Training 2.0 comes:
Enhanced data transformation and feature engineering, laying the groundwork for customization in future releases;
Support for different types of hyper-parameter tuning methods;
Enhanced model evaluation and visualization with useful metrics and charts;
Laying the groundwork for supporting new model templates; and
Added flexibility to train and serve with different configuration