PinnedInCodeXbyBakhtiyarsalah·Oct 2, 2024How can we inject our domain knowledge to the traditional ML models? Monotonic constraintsAchieving model interpretability and adherence to domain knowledge is crucial. One way to enhance both is through the implementation of…A response icon5A response icon5
PinnedInCodeXbyBakhtiyarsalah·May 21, 2024How to prevent Data Leakage problem: Illustrated with Practical ExamplesData leakage occurs when knowledge of the hold-out test set leaks into the dataset used to train the model, leading to inaccurate…A response icon1A response icon1
PinnedBakhtiyarsalah·Jan 23, 2024Comprehensive Exploration of Probability CalibrationThis blog aims to answer the following questions:
InCodeXbyBakhtiyarsalah·Nov 16, 2024Is a Model with 76% ROC score Truly Better Than One with 73%, or Just Slightly Luckier?What does it mean to have the “best model”?
InCodeXbyBakhtiyarsalah·Oct 31, 2024Can We Guide Feature Importance? How Can Key Features Be Prioritized for Greater Impact?Before diving into these questions, let’s explore why influencing feature importance can be crucial.
InCodeXbyBakhtiyarsalah·Sep 25, 2024How Does XGBoost Handle Missing Data?One of XGBoost’s notable strengths is its ability to manage missing values during training without requiring additional preprocessing or…A response icon1A response icon1
Bakhtiyarsalah·Apr 26, 2024Unveiling Data Quality Issues and Data Cleaning with Python Function: variance_analyze()Data cleaning is a critically important step in any machine learning project. The most useful data cleaning involves deep domain expertise…A response icon1A response icon1