Indeed, if one could run “pip install lib, lib.explain(model)”, why bother on the theory behind? The availability and simplicity of the methods are making them “golden hammer”. ( Interpreting Interpretability: Understanding Data Scientists’ Use of Interpretability Tools for Machine Learning) showed that not all data scientists know how to do it correctly. While others are “universal”, they could be applied to almost any model: methods such as SHAP values, permutation importances, drop-and-relearn approach, and many others.Īlthough the model’s black box unboxing is an integral part of the model development pipeline, a study conducted by Harmanpreet et al. Some of them are based on the model’s type, e.g., coefficients of linear regression, gain importance in tree-based models, or batch norm parameters in neural nets (BN params are often used for NN pruning, i.e., neural network compression for example, this paper addresses CNN nets, but the same logic could be applicable to fully-connected nets). There are a lot of ways how we could calculate feature importance nowadays. Indeed, the model’s top important features may give us inspiration for further feature engineering and provide insights on what is going on. Also, importance is frequently using for understanding the underlying process and making business decisions. Importances could help us to understand if we have biases in our data or bugs in models. Data scientists need features importances calculations for a variety of tasks.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |