Initialisation
Required Modules
Regardless of the option chosen to run B-FADE, before starting to code the main body of a script, make sure to include in its header the following:
import numpy as np
import pandas as pd
import scipy
The included modules/functions shall be utilised throughout the execution. Specifically, scipy is imported to recall probability distributions for shaping priors. numpy is required to preprocess information to define the prior parameters, in most of the envisaged cases. Whilst pandas shall be exploited to recall the file readers.
Logging
Since B-FADE implement basic logging functionalities across sub-modules, the user can easily (and optionally) customise the logging level via bfade.util.logger_manager. To do so, include the following line of code:
logger_manager(level="WARNING") # The default is "DEBUG"
where level can be selected amongst these:
DEBUGINFOWARNINGCRITICALERROR
sorted by verbosity.
Graphical Output
The user can decide on the look-and-feel appearance of the plot that B-FADE outputs. This is done via bfade.util.config_matplotlib. This function modifies the settings of matplotlib according to the user’s inputs, for instance:
config_matplotlib(font_size=12, font_family="serif", use_latex=True)