List of alternatives

Script reading the list of alternatives and identifying subsets

author:

Michel Bierlaire

date:

Mon Oct 9 10:53:03 2023

import pandas as pd
from biogeme.partition import Partition

alternatives = pd.read_csv('restaurants.dat')
alternatives

ID_COLUMN = 'ID'

all_alternatives = set(list(alternatives[ID_COLUMN]))

Set of Asian restaurants

asian = set(alternatives[alternatives['Asian'] == 1][ID_COLUMN])
print(f'Number of asian restaurants: {len(asian)}')

Set of restaurants located in downtown

downtown = set(alternatives[alternatives['downtown'] == 1][ID_COLUMN])

Set of Asian restaurants in downtown

asian_and_downtown = asian & downtown

Set of Asian restaurants, and of restaurants in downtown

asian_or_downtown = asian | downtown

Set of Asian restaurants not in downtown

only_asian = asian - asian_and_downtown

Set of non-Asian restaurants in downtown

only_downtown = downtown - asian_and_downtown

Set of restaurants that are neither Asian nor in downtown

others = all_alternatives - asian_or_downtown

def complement(a_set: set[int]) -> set[int]:
    """Returns the complement of a set"""
    return all_alternatives - a_set

Partitions.

partition_asian = Partition([asian, complement(asian)], full_set=all_alternatives)
partition_downtown = Partition(
    [downtown, complement(downtown)], full_set=all_alternatives
)
partition_uniform = Partition([all_alternatives], full_set=all_alternatives)
partition_uniform_asian = Partition([asian], full_set=asian)
partition_uniform_asian_or_downtown = Partition(
    [asian_or_downtown], full_set=asian_or_downtown
)

partitions = {
    'uniform': partition_uniform,
    'asian': partition_asian,
    'downtown': partition_downtown,
    'uniform_asian_or_downtown': partition_uniform_asian_or_downtown,
    'uniform_asian': partition_uniform_asian,
}