This repo is forked from STM, I modified some code in the repo to make it a baseline model for DRSA, which is the implementation of the model proposed in a AAAI'19 submitted paper.
Many thanks to the authors of STM
.
We have upload a tiny data sample for training and evaluation.
The full dataset for this project can be download from this link.
After download please replace the sample data in deep-bid-lands-data/
folder with the full data files.
Please install sklearn
first.
To run the demo, just execute
python demo.py
and the result(AUC, Log-Loss and ANLP) will be printed on the screen. And the result can be found in data/SurvivalModel
, the results files are named in the pattern evalution_xxxx.txt
.
To run the file with full volume data, just change the campaign_list
variable in python/demo.py
and execute it.