This repo shows some additional numerical experiments for the work on Data-driven Pricing Framework. The experiment setup and result documents can be shown in ExperimentResult.pdf and the codes are in the codes folder.
For the codes,
simulation.py shows the simulation setup for different demand models.
main_test1.py and main_test2.py are two main functions for going through all the optimization problem in one given datasets.
param.py shows the param setup for the experiment.
task_est_opt.py shows the task-based learning model. (est_opt1.py and est_opt2.py are benchmarks for linear and MNL demand models.)
e2e_network.py is the main framework for model-free learning while SigmoidNet.py and ReLuNet.py are two concrete examples.