Given a machine learning problem, the first question many people ask is usually: what is the best classification/regression algorithm to solve it? However, there is no one-size-fits-all solution or free lunch. No one could know which algorithm will work the best before trying multiple methods and fine-tuning the optimal one. We will be looking into best practices around this in the following sections.
Best practices in the model training, evaluation, and selection stage
Best practice 13 - choose the right algorithm(s) to start with
Due to the fact that there are several parameters to tune for an algorithm, exhausting all algorithms and fine-tuning each one can be extremely time-consuming and computationally expensive. We should instead short-list one to three algorithms...