Evaluation process and metrics

[jjerphan]

Accuracy is not sufficient to evaluate models. Different Metrics can be used, mainly:

• F1score
• Recall
• Precision
• AUC
• …

This issue explore the evaluation process with metrics, there is two scenarios:

• use what Keras proposes, that is its own handling of metrics as well, as the way to define custom metrics
  • this way, everything is delegated to model.compile and model.evaluate do the job
  • we may need to define more metrics
  • we will need to keep track of custom metrics somewhere for the custom_metrics argument when using keras.models.loadmodels
• or use custom evaluation process using "manual a posteriori" computations of score with the outputs of model.predict.
  • this might be more complex but we might have more liberty and flexibility for this step
  • scikit-learn proposes a bunch of metrics

[jjerphan]

Keras used to proposes those metrics before but they have been removed as they were approximated on batches. For more information see this issue.

Another package, keras-metrics proposes ready to use metrics for Keras, but it seems that there is a problem with models that get saved using model.save() — basically, metrics defined by this package aren't correctly saved/serialized in the .h5 file. See this issue.

As fchollet suggests, the best option might be to use a custom workflow (hence our second scenario).

Moving on it now! 🏃

[jjerphan]

This has been made in #23.

[jjerphan]

This has been done ine #23.
Metrics from scikit-learn has been used for the evaluation, mainly accuracy_score, precision_score, recall_score, f1_score, confusion_matrix. More may be added in the future.