Date Engineering, Statistics, AI/ML, Operations Research, Visualization
wlzpoly is a Python package that decomposes N-point wafer thickness measurements into M Zernike polynomial coefficients using LSQ or Ridge regression with LOOCV-tuned regularization. It ships with a reproducible three-stage demo (synthetic data, coefficient fitting, ground-truth verification), per-stage CLI, and a clean pure-Python API for semiconductor wafer-level process metrology.
1. Introduction: The Missing Link in Smart Manufacturing Investment in smart manufacturing and big data analytics has expanded rapidly, yet the focus has remained almost exclusively on Machine Data—the data automatically generated by equipment and systems. Despite dramatic improvements in data volume, granularity, and infrastructure, gains in defect analysis and yield improvement have fallen short…
Introduction This document classifies reproducibility problems in Python Machine Learning (ML) pipelines into three chapters, plus a fourth chapter on diagnostic techniques: This classification aligns well with the Six Sigma philosophy and toolset, and is the methodology used in semiconductor process and product development. 1. Known Problems (Observed, Identified Issues) Determinism traps recognized by standard…
This report surveys Polynomial Machine Learning (PML) at an introductory level. PML refers to the family of techniques that exploit higher-order and interaction terms of input variables to learn nonlinear relationships. The discussion is organized along three taxonomic axes and arranged into six hierarchical levels. Deep mathematical or theoretical analysis is intentionally avoided; the goal…
Thickness uniformity in thin-film deposition determines downstream yield and device performance. Variation arises along two distinct axes — within a single wafer (Within-Wafer, WiW) and across wafers over time (Wafer-to-Wafer, W2W). These two axes have different physical origins and demand different diagnostic treatments. Mixing them into a single ML target forces the model to learn…
Bottom Line Strictly speaking, no — but in practice, treat them as Out-of-Distribution (OOD). Missing-path samples in tree-based boosting models such as LightGBM, CatBoost, and XGBoost do not match the academic definition of OOD perfectly, yet they carry essentially the same risk in deployed systems. This post explains why this is a borderline case, surveys…
Machine learning (ML) models are designed under the assumption that the training distribution P_train equals the deployment distribution P_test. In reality, this assumption breaks frequently, causing sharp accuracy drops in deployed systems. This post organizes these failures into a clean taxonomy and summarizes practical mitigation strategies. 1. Hierarchy of ML Failures from train ≠ test…
This report analyzes why standard vectorization methods — statistical summary (mean/var/AUC), automatic feature extraction (tsfresh, catch22), convolutional representations (MiniRocket), and self-supervised embeddings (TS2Vec) — fail when the time series length T is very short, especially T ≤ 10. As an alternative, it argues that Raw Vectorization (Identity Mapping) is not just an easy fallback but…
1. Introduction This article summarizes how three popular gradient boosting libraries — LightGBM (Light Gradient Boosting Machine), XGBoost (Extreme Gradient Boosting), and CatBoost (Categorical Boosting) — handle missing values and unknown categories that appear during inference. We examine how the missing direction is learned during training and how each library responds to category values it…
When performing feature selection with tree-based models such as LightGBM (LGBM) or CatBoost, adding noise features to the existing set often causes truly important primary features to drop out of the importance ranking. This is not a data problem but a structural issue rooted in how tree models compete during training and how importance is…
Bondi Iceberg pool 1. Introduction: R² and Its Relation to RSQ The coefficient of determination, denoted as R² (R-squared), is one of the most widely used validation metrics in statistics and machine learning (ML) for assessing how well predicted values $\hat{y}$ agree with observed values $y_{true}$. In spreadsheet software and many engineering reports, R² is…
Reading ML papers often raises a recurring question: where does the true novelty of this paper actually live? A paper may claim a “new framework” while being a recombination of existing techniques, or may look like a “simple application” while hiding a subtle but important contribution. The 4-Axis Matrix is an analytical tool that decomposes…
