incremental

Hi, In our work, we have a large-scale logistic regression model we hate to train everyday. Is there a way we can somehow update the existing model with the new day's data? Currently, we re-train the model daily with the past ninety-day (sliding window) data. And we use the L1-regularizer...

Is there any difference between incremental f-test and F-test ??