Original Source Here
Evaluate results and establish a baseline
Now that your pipeline outputs a result, you must evaluate it with metrics.
If your problem is a classification, use Sklearn’s classification report to output precision, recall, f1, and support for each class.
If your problem is a regression, dig into error metrics like MSE, RMSE, and MAE.
No two problems can be evaluated in the same way. Always take context into account.
Now note your performance metrics along with the current state of your pipeline. Don’t be discouraged if your model performs poorly at this point.
This is your baseline. Everything you do from here on out has the goal of exceeding it.
It’s time to iterate and see if we can outperform the baseline.
Trending AI/ML Article Identified & Digested via Granola by Ramsey Elbasheer; a Machine-Driven RSS Bot