Machine Learning with Spark and Python, 2nd Edition

Index

• A
• abalone, predicting age of
  • coding sex of abalone for input to random forest regression in Python, 284–286
  • coding sex of for gradient boosting regression in Python, 280–282
  • incorporating categorical variables in PySpark model on, 213–217
  • incorporating non-numeric attributes into linear methods, 163–166
  • with PySpark ensemble methods, 317–321
  • real-valued predictions with factor variables, 50–61
• algorithms
  • bagging algorithm, 222, 235–246, 259
  • binary tree algorithms, 35–36
  • boosting algorithm, 222
  • choosing linear or nonlinear? 90–91
  • decision tree algorithm, 230
  • factors driving choices and performance of, 82–91
  • factors driving predictive algorithm performance, 89–90
  • how to decide which one to use, 11–13
  • random forest algorithm
    • bagging plus random attribute subsets, 259
    • classifying glass using, 307–310
    • described, 222
    • detecting unexploded mines with Python random forest, 292–294
    • overview, 256–259
    • performance drivers of, 260–261
• alpha values, 46, 121, 122, 124, 172, 174, 178, 179, 191, 193
• Anaconda Python distribution, 36
• area under the curve (AUC), 91, 120, 121, 122, 123, 171, 183, 188, 193, 212, 289, 291, 293, 294, 298
• argmin, 116, 136
• artificial neural nets (ANNs), 4
• attributes
  • blessed attributes, 3
  • categorical attributes, 35–36, 80, 163
  • cursed attributes, 3
  • defined, 25
  • different types of as driving modeling choices, 26–27
  • factor attributes, 80, 163
  • a.k.a.s for, 25
  • non-numeric attributes, 163–166, 279–288
  • numeric attributes, 26, 35, 61, 80, 163, 182
  • real-valued attributes, ...