Fig. 7From: Sparsifying machine learning models identify stable subsets of predictive features for behavioral detection of autismHistogram of predicted probabilities for LDA (a), logistic regression with L 2 penalization (b), and linear SVM with L 1 penalization (c) on the test set. The x-axis denotes the predicted probabilities (or decision function for SVM), and the y-axis shows the number of subjects. Each subplot corresponds to a different labelBack to article page