Imbalanced Classification with Python: Better Metrics, Balance Skewed Classes, Cost-Sensitive LearningMachine Learning Mastery, 14/01/2020 - 463 páginas Imbalanced classification are those classification tasks where the distribution of examples across the classes is not equal. Cut through the equations, Greek letters, and confusion, and discover the specialized techniques data preparation techniques, learning algorithms, and performance metrics that you need to know. Using clear explanations, standard Python libraries, and step-by-step tutorial lessons, you will discover how to confidently develop robust models for your own imbalanced classification projects. |
Índice
III Model Evaluation | 35 |
IV Data Sampling | 103 |
V CostSensitive | 177 |
VI Advanced Algorithms | 244 |
VII Projects | 303 |
VIII Appendix | 431 |
IX Conclusions | 444 |
Palavras e frases frequentes
accuracy achieved applied average balanced binary classification Brier calculate calibrated class distribution class label class weighting combination complete example configuration Consider cost cost-sensitive counter create cross-validation decision tree define model discover effective error evaluate evaluate model Example of defining Example output expected false Figure flip_y=0 function given grid search imbalance imbalanced classification dataset Imbalanced Data imbalanced dataset input Listing load the dataset logistic regression look majority make_classification(n_samples=10000 mean methods metric minority class model model n_features=2 n_redundant=0 names negative numpy import outliers oversampling performance pipeline positive positive class precision predicted probabilities predictive modeling provides random random_state=1 recall referred reports ROC AUC ROC Curve Running the example sampling scatter plot score showing skewed skill sklearn.datasets import make_classification sklearn.model_selection import SMOTE specific split standard summarize synthetic techniques threshold training dataset transform tutorial undersampling values variables yhat