How would you go about validation of a performance model?
How would you go about validation of a performance model?
Validation uses your model to predict the output in things outside your training data, and calculates the same statistical measures of work on those results. this means you need to divide your knowledge set into 2 different data files. the primary may be a training data set, that you use to get your model, while the second may be a validation data set, that you utilize to check your model’s accuracy against data you didn’t use to train the model.
Let’s say you’re creating multiple models for a project. The natural choice is to pick out the model that most accurately fits your validation data and move on. However, now we have another potential pitfall. simply because a model closely matches the validation data doesn’t mean the model matches reality. whereas the model in question performs best in this particular test, it could still be wrong.
The final step, and ultimate solution to the problem, is to match the model that performed best within the validation stage against a 3rd data set: the test data. This test data is, again, a set of the information from the original data source. It consists only of points that were used in neither the model’s development nor its validation. we tend to think about a model prepared to be used only if we tend to compare it against the test data, and the statistical calculations show a satisfactory match.
Step by step
Solved in 2 steps
