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

Ec5 revision #386

Open
wants to merge 7 commits into
base: main
Choose a base branch
from
Open

Ec5 revision #386

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
262ff93
added alternate ec5 configuration
akselx Aug 27, 2024
3190388
Added ec5 spec file name to run_setup.yaml for flexibility
akselx Aug 27, 2024
d2a4ec4
add ec5 area category job detail by sector
akselx Aug 27, 2024
0527fe9
added second alternate ec5 configuration
akselx Aug 27, 2024
3c87d6f
expanded ec5 job universe - should be parameterized in yaml soon
akselx Aug 30, 2024
c93a2e6
fixed building_hosts expression
akselx Aug 30, 2024
f5820ae
Merge remote-tracking branch 'origin/meta_changes' into ec5_revision
akselx Oct 17, 2024
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
16 changes: 10 additions & 6 deletions baus/models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,7 @@ def elcm_simulate(jobs, buildings, aggregations):


@orca.step()
def elcm_simulate_ec5(jobs, buildings, aggregations, year):
def elcm_simulate_ec5(jobs, buildings, aggregations, year,run_setup):
"""
testing docstring documentation for automated documentation creation
"""
Expand All @@ -57,7 +57,7 @@ def elcm_simulate_ec5(jobs, buildings, aggregations, year):
return

#spec_path = os.path.join("location_choice", orca.get_injectable("elcm_spec_file"))
spec_path = os.path.join("location_choice",'elcm_ec5.yaml')
spec_path = os.path.join("location_choice",run_setup.get('ec5_spec_file','elcm_ec5.yaml'))

elcm = utils.lcm_simulate(spec_path,
jobs, buildings, aggregations,
Expand All @@ -72,7 +72,7 @@ def elcm_simulate_ec5(jobs, buildings, aggregations, year):
def gov_transit_elcm(jobs, buildings, parcels, run_setup, year):

"""
This function assigns jobs in the NAICS 91 sector (Real Estate and Rental and Leasing) to vacant job spaces in transit hubs (ec5_cat='Transit_Hub').
This function assigns jobs in the NAICS 91 sector (Govt / public admin) to vacant job spaces in transit hubs (ec5_cat='Transit_Hub').

Args:
jobs (orca.DataFrameWrapper(): A pandas DataFrame containing job data.
Expand Down Expand Up @@ -114,8 +114,12 @@ def gov_transit_elcm(jobs, buildings, parcels, run_setup, year):
#buildings_df = buildings_df.rename(columns={'county_x': 'county', 'general_type_x': 'general_type'})

# Where to go? Buffers!
building_hosts = buildings_df.query('ec5_cat=="Transit_Hub" & vacant_job_spaces > 0 & general_type!="Residential"')

#building_hosts = buildings_df.query('ec5_cat=="Transit_Hub" & vacant_job_spaces > 0 & general_type!="Residential"')
building_hosts = buildings_df.loc[(buildings_df.ec5_cat.isin(['Transit_Hub', 'Transit_Hub_Low_Emp', 'Transit_Hub_Very_Low_Emp']))
& (buildings_df.vacant_job_spaces > 0)
& (buildings_df.general_type != 'Residential')
]

# first - enumerate job spaces - but index to building_id is retained
building_hosts_enum = building_hosts.index.repeat(building_hosts.vacant_job_spaces.clip(0))

Expand All @@ -125,7 +129,7 @@ def gov_transit_elcm(jobs, buildings, parcels, run_setup, year):


# second - get Move candidates
moving_jobs_candidates = jobs_df[jobs_df.sector_id.isin([91])]
moving_jobs_candidates = jobs_df[~jobs_df.sector_id.isin([11,21,48,49,31,32,33])]
print('Move candidates, by placement status: ',
moving_jobs_candidates.groupby(moving_jobs_candidates.building_id!=-1).size())
# Suppose we don't want all but just some - we could add logic for filtering later
Expand Down
8 changes: 8 additions & 0 deletions baus/summaries/core_summaries.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -151,7 +151,15 @@ def interim_zone_output(run_name, households, buildings, residential_units, parc

# ADD JOBS BY TRANSIT ZONES
jobs_by_ec5 = jobs.groupby(['zone_id','is_transit_hub']).size().unstack(1).fillna(0).astype(int)
jobs_detail_by_ec5 = jobs.groupby(['zone_id','empsix','is_transit_hub']).size().unstack([2,1])

# add sector detail
jobs_detail_by_ec5.columns = ['_'.join(map(str,col)).strip() for col in jobs_detail_by_ec5.columns.values]
jobs_detail_by_ec5 = jobs_detail_by_ec5.fillna(0).astype(int)

# merge the two in turn
zones = zones.merge(jobs_by_ec5, how='left',left_index=True,right_index=True)
zones = zones.merge(jobs_detail_by_ec5, how='left',left_index=True,right_index=True)

zones['non_residential_sqft'] = buildings.groupby('zone_id').non_residential_sqft.sum()
zones['non_residential_sqft_office'] = buildings.query('building_type=="OF"').groupby('zone_id').non_residential_sqft.sum()
Expand Down
313 changes: 313 additions & 0 deletions configs/location_choice/elcm_ec5_loose.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,313 @@
name: elcm

model_type: segmented_discretechoice

segmentation_col: empsix

choosers_fit_filters:
- building_id != -1
- sector_id==91 | sector_id==92

choosers_predict_filters:
- sector_id!=48
- sector_id!=49
- sector_id!=31
- sector_id!=32
- sector_id!=33
- sector_id!=11
- sector_id!=21

alts_fit_filters:

alts_predict_filters:
- job_spaces > 0
- ec5_cat=="Transit_Hub"

interaction_predict_filters: null

choice_column: null

sample_size: 50

estimation_sample_size: 5000

prediction_sample_size: null

default_config:
model_expression: np.log1p(non_residential_rent) + office_1500 + industrial_1500
+ retail_1500 + jobs_45 + residential_units_1500 + residential_units_45 +
ave_income_1500 + juris_ave_income + embarcadero + stanford

models:
AGREMPN:
fit_parameters:
Coefficient:
ave_income_1500: 0.1145072404482107
embarcadero: 0.0383217609446087
industrial_1500: 0.1048607860777378
jobs_45: 1.7930795280557015
juris_ave_income: 1.1538559886899389
np.log1p(non_residential_rent): 1.311505808859499
office_1500: -0.05650573556277545
residential_units_1500: -0.3949108210041755
residential_units_45: -2.0903368281528962
retail_1500: 0.03675323513418212
stanford: 0.02842426120297033
Std. Error:
ave_income_1500: 0.00892271618696153
embarcadero: 0.0014505841113969535
industrial_1500: 0.0072359190672865135
jobs_45: 0.10190979950328043
juris_ave_income: 0.035076269954942126
np.log1p(non_residential_rent): 0.014403513560735718
office_1500: 0.008903299022863714
residential_units_1500: 0.012178900647372386
residential_units_45: 0.10215972163444892
retail_1500: 0.009438819874779932
stanford: 0.0009462526563338551
T-Score:
ave_income_1500: 12.833226794273289
embarcadero: 26.418158480795547
industrial_1500: 14.491702450323404
jobs_45: 17.594770442051384
juris_ave_income: 32.895629728364675
np.log1p(non_residential_rent): 91.05457521383474
office_1500: -6.346606512672264
residential_units_1500: -32.42581842470142
residential_units_45: -20.461457751740987
retail_1500: 3.8938379608646816
stanford: 30.038765030363873
fitted: true
log_likelihoods:
convergence: -8499.27868785579
'null': -19560.115027141746
ratio: 0.5654791050020853
name: AGREMPN
FPSEMPN:
fit_parameters:
Coefficient:
ave_income_1500: 0.057703828985868774
embarcadero: 0.0011911867168234696
industrial_1500: 0.06432429919556347
jobs_45: 0.6424496319177287
juris_ave_income: 0.5297915806258436
np.log1p(non_residential_rent): 1.0976939732233044
office_1500: 0.427279957831321
residential_units_1500: -0.3058217066950211
residential_units_45: -0.4085703736869765
retail_1500: 0.015591688141929638
stanford: 0.004973029473230911
Std. Error:
ave_income_1500: 0.010054295725582854
embarcadero: 0.0018453457072225578
industrial_1500: 0.00773941948255426
jobs_45: 0.33654195821774474
juris_ave_income: 0.06178920223876191
np.log1p(non_residential_rent): 0.012963429387261138
office_1500: 0.0105368190317198
residential_units_1500: 0.01306680805685234
residential_units_45: 0.3658222544162435
retail_1500: 0.011422776146425341
stanford: 0.001173798076018263
T-Score:
ave_income_1500: 5.739221379677854
embarcadero: 0.6455087044997833
industrial_1500: 8.311256333961415
jobs_45: 1.9089733574975511
juris_ave_income: 8.57417738747713
np.log1p(non_residential_rent): 84.6762026028377
office_1500: 40.55113374777029
residential_units_1500: -23.4044691989369
residential_units_45: -1.1168548899217403
retail_1500: 1.3649648686154916
stanford: 4.236699288262878
fitted: true
log_likelihoods:
convergence: -8537.978007402904
'null': -19560.115027141746
ratio: 0.5635006238176233
name: FPSEMPN
HEREMPN:
fit_parameters:
Coefficient:
ave_income_1500: 0.026120422638807334
embarcadero: -0.0007530248295755161
industrial_1500: -0.05364340018896362
jobs_45: 0.6023382363503588
juris_ave_income: -0.33411148862818973
np.log1p(non_residential_rent): 1.2047384438218778
office_1500: 0.2814964183304126
residential_units_1500: -0.20386741728999144
residential_units_45: -0.5380536161562809
retail_1500: 0.05478483429770649
stanford: 0.004822373073213551
Std. Error:
ave_income_1500: 0.01179650033013969
embarcadero: 0.0017797725323409328
industrial_1500: 0.007393490887050828
jobs_45: 0.29216141714255633
juris_ave_income: 0.055002982874583835
np.log1p(non_residential_rent): 0.012312234227100198
office_1500: 0.010921535249383276
residential_units_1500: 0.014504011403046668
residential_units_45: 0.3157844224643049
retail_1500: 0.011235361163642243
stanford: 0.0011146463126104442
T-Score:
ave_income_1500: 2.214251846547274
embarcadero: -0.42310172558122544
industrial_1500: -7.255490134290449
jobs_45: 2.0616624954843226
juris_ave_income: -6.074424897828193
np.log1p(non_residential_rent): 97.84888929177075
office_1500: 25.77443664308169
residential_units_1500: -14.055933329393804
residential_units_45: -1.7038637053640622
retail_1500: 4.8761079861848
stanford: 4.32637063313815
fitted: true
log_likelihoods:
convergence: -9798.308660164865
'null': -19560.115027141746
ratio: 0.49906692028300115
name: HEREMPN
MWTEMPN:
fit_parameters:
Coefficient:
ave_income_1500: 0.027210485174825445
embarcadero: 0.016062177732133746
industrial_1500: 0.309189196732516
jobs_45: 1.0629885427323065
juris_ave_income: -0.3434887510642488
np.log1p(non_residential_rent): 1.2968958396833719
office_1500: 0.08221294007271106
residential_units_1500: -0.34553582926132437
residential_units_45: -0.566697351210849
retail_1500: -0.09582624645612992
stanford: 0.006640676090924627
Std. Error:
ave_income_1500: 0.007800174076015303
embarcadero: 0.002062414018164512
industrial_1500: 0.008107424606967483
jobs_45: 0.4093027394102561
juris_ave_income: 0.06560879025736568
np.log1p(non_residential_rent): 0.01698378199243969
office_1500: 0.010497341736661427
residential_units_1500: 0.012153265752086521
residential_units_45: 0.43891865236043187
retail_1500: 0.011089187887431805
stanford: 0.001354002703906375
T-Score:
ave_income_1500: 3.488445887187923
embarcadero: 7.788047206170861
industrial_1500: 38.13654911657152
jobs_45: 2.5970716547461024
juris_ave_income: -5.235407476907204
np.log1p(non_residential_rent): 76.36083884382663
office_1500: 7.831786573698614
residential_units_1500: -28.431520902272823
residential_units_45: -1.2911215965948233
retail_1500: -8.641412466708845
stanford: 4.904477717633722
fitted: true
log_likelihoods:
convergence: -6876.87407643065
'null': -19560.115027141746
ratio: 0.6484236382614186
name: MWTEMPN
OTHEMPN:
fit_parameters:
Coefficient:
ave_income_1500: 0.003758015652852145
embarcadero: 0.002823274243760113
industrial_1500: 0.11431769687781299
jobs_45: 0.8883398274765266
juris_ave_income: 0.13546788186472114
np.log1p(non_residential_rent): 1.1586137002160306
office_1500: 0.32976290183260726
residential_units_1500: -0.21661959316114188
residential_units_45: -0.6970395454883818
retail_1500: -0.05831158133376016
stanford: 0.00883928249971089
Std. Error:
ave_income_1500: 0.00925104003798784
embarcadero: 0.0019358907852154452
industrial_1500: 0.007426647094298184
jobs_45: 0.34968124426110947
juris_ave_income: 0.05482050608539363
np.log1p(non_residential_rent): 0.01286841271512767
office_1500: 0.01046853731045222
residential_units_1500: 0.012775949221261509
residential_units_45: 0.370390239408811
retail_1500: 0.011278778826708913
stanford: 0.0012706586559210172
T-Score:
ave_income_1500: 0.40622628779256015
embarcadero: 1.4583850831470904
industrial_1500: 15.39290818942784
jobs_45: 2.5404274380046443
juris_ave_income: 2.4711169512682627
np.log1p(non_residential_rent): 90.03547880104931
office_1500: 31.500379857590836
residential_units_1500: -16.955264098940482
residential_units_45: -1.8819058153393668
retail_1500: -5.170026137552621
stanford: 6.9564571559179855
fitted: true
log_likelihoods:
convergence: -9041.43091307284
'null': -19560.115027141746
ratio: 0.537761874072474
name: OTHEMPN
RETEMPN:
fit_parameters:
Coefficient:
ave_income_1500: 0.07536262083097184
embarcadero: 0.009422767487277467
industrial_1500: 0.02290574058702665
jobs_45: 0.004339163785697172
juris_ave_income: -0.0037676081058425847
np.log1p(non_residential_rent): 1.2629449388166134
office_1500: -0.012353404839634766
residential_units_1500: -0.3093391456455967
residential_units_45: 0.31921212359055945
retail_1500: 0.3160661700215789
stanford: 0.012506607844302296
Std. Error:
ave_income_1500: 0.010592203515181973
embarcadero: 0.0017603384834679599
industrial_1500: 0.007088616303727654
jobs_45: 0.2661154301000682
juris_ave_income: 0.05813951958221012
np.log1p(non_residential_rent): 0.013106202027149617
office_1500: 0.01022345631152455
residential_units_1500: 0.0129051095090393
residential_units_45: 0.2908385113571614
retail_1500: 0.010770370905234333
stanford: 0.0010760030146218305
T-Score:
ave_income_1500: 7.114914354029678
embarcadero: 5.352815708893734
industrial_1500: 3.2313415771962894
jobs_45: 0.01630557004554566
juris_ave_income: -0.06480287647570139
np.log1p(non_residential_rent): 96.3623890582803
office_1500: -1.2083393779175442
residential_units_1500: -23.97028443880481
residential_units_45: 1.0975579612926643
retail_1500: 29.345894658834144
stanford: 11.623208926322421
fitted: true
log_likelihoods:
convergence: -9381.605218920939
'null': -19560.115027141746
ratio: 0.5203706519157499
name: RETEMPN

fitted: true

choice_mode: aggregate

remove_alts: true

probability_mode: single_chooser
Loading