LF_EvaluateClassification Processing Resource

The classification evaluation PR allows you to obtain a fast accuracy score for a classification problem. This is a helpful place to start in tuning your choice of algorithm, features and parameters. However for a kappa statistic, you will need to create separate training and test corpora, and train and apply using training and application PRs, then use Corpus QA. The model isn’t saved when you use the evaluation PR. This PR has no init time parameters. Here are the runtime parameters:

For an example feature specification, see the classification training PR documentation.