A track evaluate a model's performance.
The following metrics are recorded:
Time, which should be interpreted with wisdom. Indeed time can depend on the architecture
and local resource situations. Comparison via FLOPS should be preferred. - The model's memory footprint.
The model's predictive performance on the track's dataset.
Type → str
The name of the track.
The datasets that compose the track.
The metric(s) used to track performance.