Figure 1From: Prediction of the potency of mammalian cyclooxygenase inhibitors with ensemble proteochemometric modeling Ensemble modeling with model stacking. A. A set of models are trained with diverse machine learning algorithms (Model1.. Model n in the Figure). The predictions of these models on each datapoint in the training set calculated during cross validation, are used as descriptors to create a new training matrix, which rows are indexed by the datapoints in the training set and columns by the models in the library. A machine learning model is trained on this matrix. The resulting model is the model ensemble. B. The model ensemble is then applied on the test set.Back to article page