The less common label in aclass-imbalanced dataset . For example, given a dataset containing 99% negative labels and 1% positive labels, the positive labels are the minority class.1
목차
Notes
A training set with a million examples sounds impressive. However, if the minority class is poorly represented, then even a very large training set might be insufficient. Focus less on the total number of examples in the dataset and more on the number of examples in the minority class.1
If your dataset doesn’t contain enough minority class examples, consider using downsampling to supplement the minority class.1
See also
Footnotes
목차 관련 글 Recall (machine learning) “Recall is a much more useful metric for class-imbalanced datasets than accuracy.” Class (machine learning) “In a multi-class classification model that identifies dog breeds, the classes might be poodle, beagle, pug, and…” Precision (machine learning) 관련 절: When (not) to use Introduction to Machine Learning 관련 절: Definition of [[Machine learning|ML]] ML crash course - Classification “When the total of actual positives is not close to the total of actual negatives, the dataset is imbalanced.” Positive class (machine learning) “For example, the positive class in a cancer model might be "tumor."” Accuracy (machine learning) 관련 절: When to use Self-training (machine learning) “Use the model created in Step 1 to generate predictions (labels) on the unlabeled examples, moving those in whi…” 분류 모델 “A model that predicts the positive or negative class for a particular medical condition.” Machine Learning Systems “Machine Learning Systems provides a systematic framework for understanding and engineering machine learning (ML…” Epoch (machine learning) “An epoch represents N/batch size training iterations, where N is the total number of examples.” Dataset (machine learning) “A dataset is characterized by its size and diversity.” Generalization (machine learning) “To encourage generalization, regularization helps a model train less exactly to the peculiarities of the data i…” Classification threshold “If this raw value is less than the classification threshold, then the negative class is predicted.” Negative class (machine learning) “In binary classification, one class is termed positive and the other is termed negative.” 지도 학습 “Supervised machine learning is analogous to learning a subject by studying a set of questions and their corresp…” Iteration (machine learning) “A backward pass (backpropagation) to adjust the model's parameters based on the loss and the learning rate.” Semi-supervised learning “Semi-supervised learning can be useful if labels are expensive to obtain but unlabeled examples are plentiful.” 배치 크기 “For instance, if the training set contains a million examples, then the batch size would be a million examples.” 기계학습 “2024-08-24 - What's Really Going On in Machine Learning?”