Leave-some-out cross-validation

{
  "id": "MLC-T-CHM-006505",
  "term": "Leave-some-out cross-validation",
  "field": "Chemistry",
  "definition": "A statistical method for evaluating the predictive performance of a model by systematically removing a subset of data points from the training set, building the model with the remaining data, and then predicting the values for the removed subset. This process is repeated multiple times, with different subsets held out, to obtain a more reliable estimate of the model's generalization ability. The size of the 'some' subset can vary, from a single data point (leave-one-out) to a larger fraction of the dataset.",
  "subject": "Chemistry and Human Health",
  "references": [
    "PAC, 2016, 88, 239. 'Glossary of terms used in computational drug design, part II (IUPAC Recommendations 2015)' on page 249 (https://doi.org/10.1515/pac-2012-1204)"
  ],
  "url": "https://mlchart.com/terminology/chemistry/leave-some-out-cross-validation/"
}