How to choose a good baseline to know whether your model is performing well or not?
Baselines - ML Projects
- A baseline is a model that is both simple to set up and has a reasonable chance of providing decent results. It gives you a lower bound on expected model performance.
- Your choice of a simple baseline depends on the kind of data you are working with and the kind of task you are targeting.
- You can look for external baselines such as business and engineering requirements, as well as published results from academic papers that tackle your problem domain.
- You can also look for internal baselines using simple models and human performance.
- There is a tradeoff between cost and quality when designing human baselines. More specialized domains require more skilled labelers, so you should find cases where the model performs worse and concentrate the data collection effort there.