No matter where you data are located on the map, each time a kepler.gl map is visualised, the starting location is by default San Francisco.
For example:
import pandas as pd
from keplergl import KeplerGl
df = pd.read_csv("<dataset.csv>")
# df contains data in France
map_1 = KeplerGl(height=400, data={"data_1": df}, config=config)
map_1
map_1 will look like:
Which means that you have to manually pan to where the uploaded data are in the map (to France in this case).
Is there a way to change the default starting point to where the data are located?

I found the solution shortly after posting the question: The initial coordinates can be embedded in the config dictionary:
Note that if the config dictionary is defective, then the value will be set to the standard ones with no warnings raised.