search

tkuo-tkuo / DeepMutationOperators

Developed with the UROP, Detecting Deep Learning Software Defects (Spring 2019), HKUST
20 stars 11 forks source link

How to run the code/tool #1

Open Abdullawal opened 4 years ago

Abdullawal commented 4 years ago

Hey KuoTzu, can you help to explain or guide on how to run the files code please? This seems to be exactly the codes/tool I need to start my thesis project. I'm new to Deep Learning and Mutation testing topics. Please guide me on how to go about running and making use of the code/tool

tkuo-tkuo commented 4 years ago

Sure, you can start with running two demo notebooks: example_model_level.ipynb and example_source_level.ipynb. Please let me know if you encounter any issue running these two notebooks.

Abdullawal commented 4 years ago

Hey, i run the first two notebooks (xx_model and xx_source_level) and all executed well. Im starting to understanding the implementation now. Whats the next step now to complete full running of the tool and to have or save the ratios/numbers to the h5 file which I can use later for justification and evaluation. Thanks

tkuo-tkuo commented 4 years ago

Hey, code related to model (instantiation, training, save, load), please refer to the files _model_generators.py. For storing a model, you may want to look at the save_model** function in network.py. As for the ratios/numbers, there should be many approaches to do so (record your experimental data). Keep in touch and let me know if there anything I could help with!