Skip to content

Avoid Inf Loop in Joint Tour Participation #62

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
dhensle wants to merge 4 commits into
base: BayDAG
Choose a base branch
from
Open

Avoid Inf Loop in Joint Tour Participation #62

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
08b30c7
avoid infinite loop if forcing does not work
dhensle Mar 21, 2023
a732ee6
better error message
dhensle Mar 30, 2023
ccbdc6d
formatting
dhensle Mar 30, 2023
9b34aae
additional documentation
dhensle Jul 27, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 13 additions & 2 deletions activitysim/abm/models/joint_tour_participation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,8 +132,12 @@ def participants_chooser(probs, choosers, spec, trace_label):
and then check to see if the tour statisfies this requirement, and rechoose for any that
fail until all are satisfied.

In principal, this shold always occur eventually, but we fail after MAX_ITERATIONS,
just in case there is some failure in program logic (haven't seen this occur.)
In principal, this shold always occur eventually, but we fail after MAX_ITERATIONS.
Failure can happen due to sampling on really small probabilities of participation for
household members. There is an optional setting to force participation of all household
members with non-zero probability after MAX_ITERATIONS. Even with this forced participation,
tours can still be unsatisfied due to logical consistencies in upstream models or bad input
data in estimation mode. If forced participation does not work, the run will crash.

The return values are the same as logit.make_choices

Expand Down Expand Up @@ -209,6 +213,13 @@ def participants_chooser(probs, choosers, spec, trace_label):
probs[choice_col] = np.where(probs[choice_col] > 0, 1, 0)
non_choice_col = [col for col in probs.columns if col != choice_col][0]
probs[non_choice_col] = 1 - probs[choice_col]
if iter > MAX_ITERATIONS + 1:
raise RuntimeError(
f"{num_tours_remaining} tours could not be satisfied even with forcing participation."
" This is likely due to some inconsistency between the tour frequency & composition model"
" and the joint tour participation model, e.g. overlapping time windows with mandatory tours."
" It could also be inconsistencies in the data if run in estimation mode."
)
else:
raise RuntimeError(
f"{num_tours_remaining} tours could not be satisfied after {iter} iterations"
Expand Down