Skip to content
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

Jump to bottom

Location and modechoice logsum written out in estimation mode #898

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Location and modechoice logsum written out in estimation mode #898

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
57 changes: 19 additions & 38 deletions activitysim/abm/models/location_choice.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -517,10 +517,13 @@ def run_location_sample(
["person_id", "alt_dest", "prob", "pick_count"]
].set_index("person_id")
choices = choices.append(new_choices, ignore_index=False).sort_index()
# making probability the mean of all other sampled destinations by person
# FIXME is there a better way to do this? Does this even matter for estimation?
choices["prob"] = choices["prob"].fillna(
choices.groupby("person_id")["prob"].transform("mean")
# making prob 0 for missing rows so it does not influence model decision
choices["prob"] = choices["prob"].fillna(0)
# sort by person_id and alt_dest
choices = (
choices.reset_index()
.sort_values(by=["person_id", "alt_dest"])
.set_index("person_id")
)

return choices
Expand Down Expand Up @@ -844,41 +847,14 @@ def run_location_choice(
)
estimator.write_override_choices(choices_df.choice)

if want_logsums:
# if we override choices, we need to to replace choice logsum with ologsim for override location
# fortunately, as long as we aren't sampling dest alts, the logsum will be in location_sample_df

# if we start sampling dest alts, we will need code below to compute override location logsum
assert estimator.want_unsampled_alternatives

# merge mode_choice_logsum for the overridden location
# alt_logsums columns: ['person_id', 'choice', 'logsum']
alt_dest_col = model_settings.ALT_DEST_COL_NAME
alt_logsums = (
location_sample_df[[alt_dest_col, ALT_LOGSUM]]
.rename(columns={alt_dest_col: "choice", ALT_LOGSUM: "logsum"})
.reset_index()
)

# choices_df columns: ['person_id', 'choice']
choices_df = choices_df[["choice"]].reset_index()

# choices_df columns: ['person_id', 'choice', 'logsum']
choices_df = pd.merge(choices_df, alt_logsums, how="left").set_index(
"person_id"
)

logger.debug(
f"{trace_label} segment {segment_name} estimation: override logsums"
)

if state.settings.trace_hh_id:
estimation_trace_label = tracing.extend_trace_label(
trace_label, f"estimation.{segment_name}.survey_choices"
)
state.tracing.trace_df(choices_df, estimation_trace_label)

if want_logsums & (not skip_choice):
# choices_df currently has location choice logsums in it
# do not need to override location choice logsum since it is alts dependent, not choice dependent
# choices_df:
# person_id choice logsum
# 1875722 13 10.627001
# 2159058 9 7.544127

# grabbing index, could be person_id or proto_person_id
index_name = choices_df.index.name
# merging mode choice logsum of chosen alternative to choices
Expand All @@ -895,6 +871,11 @@ def run_location_choice(
.drop(columns=model_settings.ALT_DEST_COL_NAME)
.set_index(index_name)
)
# choices_df now has mode choice logsum of override alternative:
# person_id choice logsum mode_choice_logsum
# 1875722 13 10.627001 3.526035
# 2159058 9 7.544127 0.372482
# 3188484 13 9.312651 3.179640

choices_list.append(choices_df)

Expand Down
Loading