If a visualization should explain the outcomes of a single computation (e.g., for simple black-scholes):
Define an input class for the functionality (i.e., what parameters/inputs will be coming from the frontend)
Create a method that takes in all parameters needed to compute a solution (can be a combination of user inputs and other parameters)
Additionally, if the visualization should display the change of a computation across an entire exemplary time-series (e.g., the change of a price based on a changing underlying):
Optional: Place any example data files in data/ (a simple CSV with one column works best, but it is also possible to add multi-column data files) and import the data using pd.read_csv
Create functionality that applies the "single-datapoint computation" across the entire series, combining the series with fixed inputs (e.g., varying strike price but otherwise fixed black-scholes parameters)
Return the resulting series as a result. The frontend can then take all values of the result series and display them on (potentially) different axes.