# Evaluation

*Avalanche* offers significant support for *defining your own eveluation protocol* (classic or custom metrics, when and on what to test). You can find **examples** related to the benchmarks here: 

* [Eval Plugin](https://github.com/ContinualAI/avalanche/blob/master/examples/eval_plugin.py): *this is a simple example on how to use the Evaluation Plugin (the evaluation controller object)*
* [Standalone Metrics](https://github.com/ContinualAI/avalanche/blob/master/examples/standalone_metric.py): *how to use metrics as standalone objects.* 
* [Confusion Matrix](https://github.com/ContinualAI/avalanche/blob/master/examples/confusion_matrix.py): *this example shows how to produce confusion matrix during training and evaluation.*
* [Dataset Inspection](https://github.com/ContinualAI/avalanche/blob/master/examples/dataset_inspection.py)*: this is a simple example on how to use the Dataset inspection plugins.*
* [Mean Score](https://github.com/ContinualAI/avalanche/blob/master/examples/mean_scores.py): *example usage of the mean\_score helper to show the scores of the true class, averaged by new and old classes.*
* [*Task Metrics*](https://github.com/ContinualAI/avalanche/blob/master/examples/task_metrics.py)*: this is a simple example on how to use the Evaluation Plugin with metrics returning values for different tasks.*