Expand description
The metric module.
Modules§
- State module.
- Module responsible to save and exposes data collected during training.
Structs§
- The accuracy metric input type.
- The accuracy metric.
- Memory information
- CPU Temperature in celsius degrees
- General CPU Usage metric
- Track basic cuda infos.
- The hamming score, sometimes referred to as multi-label or label-based accuracy.
- The hamming score input type.
- Track the learning rate across iterations.
- The loss metric input type.
- The loss metric.
- Data type that contains the current state of a metric at a given time.
- Metric metadata that can be used when computing metrics.
- The top-k accuracy metric input type.
- The Top-K accuracy metric.
Enums§
- Numeric metric entry.
Traits§
- Adaptor are used to transform types so that they can be used by metrics.
- Metric trait.
- Declare a metric to be numeric.
Functions§
- Format a float with the given precision. Will use scientific notation if necessary.