How can I test the accuracy of my ordinary least squares model?
Is it a simple comparison between the predicted values of my test set and their actual values (with perhaps a maximum threshold of epsilon -> 0.00001)?
I'm not sure if this is the best approach in terms of accuracy.
Accuracy as in TP+TN/P+N is used in classification settings; because there you have correct labels and false labels and you want to know how well you are classifying. Since you are using OLS I assume you are in a regression setting. Something like MSE will be better suited I think. Are you in a classification setting with OLS?