Learn how to use Gramian Angular fields to improve performance on high frequency datasets:
Traditional feature engineering methods like time aware aggregation and spectrograms can have limitations. Spectrograms cannot capture correlation between each segment of the signal with other segments of the signal. If we try to do this with tabular aggregates it becomes a high dimensionality problem.
About this Accelerator
This example notebook shows how to generate advanced features other than traditional aggregated features used for high frequency data use cases. The notebook uses Gramian Angular Fields to generate features from high frequency time series datasets like sensor readings. These GAF features can be used with DataRobot Visual AI. This approach of feature engineering can be used as an augmentation to other feature engineering techniques like spectrograms.
GAF representation of two different signals
This notebook illustrates how to generate GAF interpretations of the signal.