Runtime considerations are often important for many applications. Typically you should look at training time and prediction time for an ML algorithm.
Some common questions to ask include:
- Training: Do you want to train the algorithm in a batch mode? How often do you need to train? If you need to retrain your algorithm every week, you are OK to choose something that might take significant training time. If you need to keep re-running your algorithm every 10 seconds to train with new data, the training time becomes very important.
- Prediction: What is the SLA during prediction ? Some algorithms have a fast prediction SLA. You need to use an algorithm and a prediction framework that supports a fast SLA.