For each point we have the predicted class / regression value in column response. If the learner predicts probabilities, a column ".prob.response" is added that contains the probability of the predicted class

predict_grid(learners, task, grid_points, expand_range)

## Arguments

learners list of trained learners, each learner belongs to one resampling iteration the task all learners are trained on (int): see sequenize see sequenize