The report generator is devoted to producing a visual digest of the utility and privacy performance of the synthetic dataset under test.
The function generates a Streamlit dashboard with graphs and tables displaying the metrics computed in the previous modules.
Every result in output from the previous utility and privacy modules is saved in a JSON file. This file is loaded into the Streamlit app and graphically represented to provide a visual summary of the performance of the synthetic dataset under test.
report()The function takes as input the original versions of the real and the synthetic dataset, before being pre-processed. They are used to plot the distribution of the features.
It is possible to specify the path to the JSON file where the information computed with the other functions was stored.
If the path to the JSON file is not specified, the function will look for the JSON file in the directory where you are currently working.
Make sure to consistently provide the same path to the JSON file, so that all previously computed information is stored in a single file.
# Searches for the JSON file in the current directory
report(real_data, synth_data)
# Searches for the JSON on ffile in path/to/json
report(real_data, synth_data, "path/to/json")
After running the function report(), click on the link provided to open the report in your browser.