AbstractThe proper hydraulic design of microirrigation system subunits requires the characterization of minor losses. To this end, machine learning models based on artificial neural networks [multilayer perceptron (MLP)], support vector machines [support vector regression (SVR)], and an ensemble of decision trees [extreme gradient boosting (XGB)] were developed and validated to predict minor losses caused by fittings commonly used in microirrigation subunits. The databases for learning are collections of experiments with commercial fittings classified as I, Y, and T. The features considered were fluid properties along with geometric and operational characteristics. Semiempirical models based on dimensional analysis were less accurate than machine learning–based models. The MLP model presented the best performance for the evaluated processes, although it requires a considerable amount of data and an extensive calibration of the hyperparameters. The SVR model was predominantly more appropriate based on the radial basis function. However, it is computationally expensive, and the estimator may be more compromised by noise. The XGB model achieved the lowest computational cost and provided good accuracy with the test set but was less related to the theoretical power-law function expected in these hydraulic phenomena. An open-source web application was developed to support the use and comparison of the models; it can serve as an online tool for the design and simulation of minor losses.