I'm running some simple classifiers in Weka to test the waters, and I can't find this particular piece of output data.
When using something like a % split, how can I obtain the training set accuracy, and not just the testing set accuracy?
You use training data to build your model, so it does not make sense at all because you will get a high % of accuracy. And you cannot guide the "fine tuning" with these results because you can get overfitting (do not generalize).
But anyway, you can use the training set to build your model, save it. And once you have it, use your training set as test set with the same model.
You can also click "More Options" and "Output predictions" to know what happened in detail with each instance.