Hi, I’m currently testing TimberStrike on the Credit Card Fraud dataset (284k samples):
https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud
After running for about 5 days, the code still hasn’t completed attacking the first tree. I’d like to ask whether this behavior is expected for a dataset of this size.
I’m running without a Gurobi at the moment (I don't have a license). Could you share how much performance improvement is typically observed when using Gurobi (e.g., rough speedup or feasibility differences)? Thanks!
Hi, I’m currently testing TimberStrike on the Credit Card Fraud dataset (284k samples):
https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud
After running for about 5 days, the code still hasn’t completed attacking the first tree. I’d like to ask whether this behavior is expected for a dataset of this size.
I’m running without a Gurobi at the moment (I don't have a license). Could you share how much performance improvement is typically observed when using Gurobi (e.g., rough speedup or feasibility differences)? Thanks!