deepice_GAN is a project aim at constructing fake data in order to solve the problem of Cardinality Estimation. deepice_GAN.py is the main python code file.
Type in python deepice_GAN_.py and you will start the experiment environment.
- deepice_GAN.py: the main python file
- get_query.py: build up query workload
- workload_generator_cat.py: get all category combination
- workload_generator_num.py: get possible numeric combination of sample_size
- get_query_pro.py build up query workload pro
- DCGAN3_model.py: train the model for generating fake data
- generator.py: generate fake data
- transformer.py: data transformer