Progress bar for sklearn model that does NOT have a "verbose" parameter

user_5794476 0
Viewed 668

When I'm training a machine learning model I like to know an estimate of how long it will take to finish running. If I'm using the sci-kit learn library, and presuming that I do not have a loop where I can use something like tqdm, is there a way to gauge the progress of my model?

I know some sci-kit learn models have a "verbose" parameter but there are many that do not, such as AdaBoostClassifier, LinearRegression and OneVsRestClassifier.

What would you recommend using in these cases?

scikit-learn machine-learning progress-bar python
1 Answers

I am not sure how a progressbar should work. E.g. LinearRegression can be implemented as an Ordinary Least Squares Proplem which is the computation of Pseudo Inverse. For AdaBoostPaper you loop once about each each estimator so maybe you can estimate your training duration by timing the fit of few estimators and then scale it up?

user_90172590
Related