Google interview question
classification vs regression metrics for evaluation how to handle missing, corrupt data segmentation Objective/loss function definitions how do you imagine an ML system, broadcasting in numpy ?
Interview Answer
Anonymous
13 Aug 2019
# KNN implementation # Input data.shape = (1000,10) labels = (1000) Observation = (1,10) K = 5 (1,0,0,1,0,0,1,0,0,0) (1,0,0,0.5,0,0,1,0,0,0) # sqrt((x1-xo1)^2) + (x2-xo2)^2) #for each i, i is in range(0,10) #remove def something_to(): dist = srqt(((data-data_point)*(data-data_point)).sum(axis=1)) Sorted_distances = dist.sort() Top_k = sorted_distances[:k]
1