PinnedAnil OzturkSetting-Up Kaggle Environment in few LinesHey there! I have just started to write blog posts about the tricks I would like to share about data science and machine learning. I will…Oct 16, 20222Oct 16, 20222
Anil OzturkQuadratic Weighted Kappa (QWK) Metric and How to Optimize ItAccuracy, precision, recall, and F1 score are commonly used metrics for most of the classification problems. But some specific scenarios…Jul 19Jul 19
Anil OzturkNested Cross-Validation Against OverfittingIn machine learning tasks, we check our models with a validation set so that they do not overfit. In fact, we use the cross-validation…Dec 4, 2022Dec 4, 2022
Anil OzturkSpeeding up I/O: Parquet and FeatherSome of our problems consist of data we read from local storage. Read-process-write operations can be comfortable ,n relatively small…Nov 27, 2022Nov 27, 2022
Anil OzturkF-Beta: Weighting Precision and RecallWe are using some standard metrics / evaluation functions to get an insight on robustness and reliability of our classifier models. The…Nov 15, 2022Nov 15, 2022
Anil OzturkAdversarial Validation: a Sanity Checker and an ExploiterIdeally, we would expect our training and test data to come from similar distributions. However, the opposite can happen in some real-life…Oct 30, 20221Oct 30, 20221
Anil OzturkStratification on Regression ProblemsHi! In this article I am going to try to make an example on how to generate splits on regression problems with preserving the…Oct 23, 20221Oct 23, 20221