Add Plots to the Figure

In plots of Tunny's optimization results, do you sometimes want to plot any value, not just the optimization results? For example, you may want to compare the pre-optimization values with the optimization results.
Unfortunately, this operation cannot be done on Grasshopper, but it can be done using Python code.

This operation can be achieved with a few additions to the code that displays the optimization results using the code introduced in the previous section.
First, read the optimization results and create a fig. This fig is a graph object in a library called plotly.

import optuna

# set storage path
storage = "sqlite:///PATH_TO/Tunny_Opt_Result.db"

# load tunny result RDB file to get study data
study = optuna.load_study(storage=storage, study_name="your_study_name")

# get figure instance
fig = optuna.visualization.plot_pareto_front(study)

In the above example, the scatter plot is being created, so any values are added using add_scatter3d, which adds values to the scatter plot. After add values, the plot is displayed and saved below. The following example adds a new point at coordinates [1,1,1] with the name "added point" displayed on mouse hover.

# add data point
fig.add_scatter3d(x=[1], y=[1], z=[1], name="added point")

# show figure
fig.show()

# save figure as html
fig.write_html(./fig.html)