Mandatory tour generation model

activitysim/defaults/variables.py

@@ -20,6 +20,11 @@ def income_segment(households):
  labels=[1, 2, 3, 4])
 
 
+@sim.column("households")
+def non_workers(households, persons):
+ return persons.household_id.value_counts() - households.workers
+
+
 @sim.column("households")
 def drivers(households, persons):
  # we assume that everyone 16 and older is a potential driver

example/README.md

@@ -0,0 +1,64 @@
+This is a list of items to double check before using in practice:
+
+* Make sure the units in things like distance_to_work match the walk thresholds
+ in the mandatory tour frequency spec. The original divided by 100. This is
+ true also of round trip auto to work and round trip auto to school.
+
+* There might be a few variables left off of some of the models. Look for 
+`head` in reading of the spec files as this is meant to eliminate some of the
+ rows. Also can look for `#` to comment out variables in the spec.
+
+* Go back to the 3 school location choices, and run the models for the 
+appropriate persons.
+
+* Probably needs code review of the variable definitions. How much of the 
+variable definitions are shared between regions and how much unique? Age 
+categories are shared? Income categories are unique?
+
+
+
+
+A few overarching principles
+
+* A little discussion of "NOT so object oriented" - this is more like a 
+database - data is in standard tables, NOT in objects
+
+* The implications of this are that most of the core code is pandas and thus 
+the quality is controlled by the larger community. We are thankful that its 
+quality is very high. Specifically, there's not so much code in activitysim 
+"proper"
+
+* What it takes to add a new model
+ * define a new model 
+ * define any new data sources necessary 
+ * add any new assumptions in settings.yaml
+ * co-create the spec and any variables that are too complicated (or 
+ reusable) for the spec
+ * run in notebook
+
+* Literally everything is really Python functions that compute something. 
+Case study of `num_under16_not_at_school` to show the inter-dependencies.
+
+
+
+
+A few questions about "best practices"
+
+* What to put into the default data sources and variable specs and what to 
+put in the example / client-specific stuff?
+
+* Want to split up injectables from variables from tables or all one big file
+ so it's easier to search?
+
+* How much variable computation to put in excel versus Python
+
+* There were some hard coded limits in the original csv - (area_type < 4 and 
+distance_to_work < 3) - these are now just left in the csv spec. Why would 
+this be different than (income_in_thousands > 50)? I've made an effort to 
+not have such "magic numbers" in Python code. (Elizabeth: MAX_NUM_AUTOS 
+exists now) 
+
+* Want to name or number the person types in the spec files?
+
+* Testing for client-specific code? It's harder because outputs are "data 
+dependent." It's easier to take a small dataset and make sure it always runs.

example/configs/auto_ownership.csv

@@ -0,0 +1,30 @@
+Description,Expression,cars0,cars1,cars2,cars3,cars4
+2 Adults (age 16+),drivers==2,,0,3.0773,3.1962,2.6616
+3 Adults (age 16+),drivers==3,,0,3.5401,5.5131,5.208
+4+ Adults (age 16+),drivers>3,,2.0107,6.3662,8.5148,9.5807
+Persons age 16-17,num_adolescents,,0,-0.881,-1.7313,-1.7313
+Persons age 18-24,num_college_age,,-0.4087,-1.0095,-1.0107,-1.0107
+Persons age 35-34,num_young_adults,,0,-0.4849,-0.8596,-0.8596
+Presence of children age 0-4,num_young_children>0,,0.3669,0.7627,0.7627,0.7627
+Presence of children age 5-17,(num_children+num_adolescents)>0,,0.0158,0.2936,0.4769,0.4769
+"Number of workers, capped at 3",@df.workers.clip(upper=3),,0,0.2936,0.6389,0.8797
+"Piecewise Linear household income, $0-30k","@df.income_in_thousands.clip(0, 30)",,0.0383,0.054,0.0559,0.0619
+"Piecewise Linear household income, $30-75k","@(df.income_in_thousands-30).clip(0, 45)",,0,0.0083,0.011,0.0147
+"Piecewise Linear household income, $75k+, capped at $125k","@(df.income_in_thousands-75).clip(0, 50)",,0,0.0083,0.011,0.0147
+"Density index up to 10, if 0 workers","@(df.workers==0)*df.density_index.clip(0, 10)",,0,-0.2028,-0.3654,-0.3654
+"Density index in excess of 10, if 0 workers",@(df.workers==0)*(df.density_index-10).clip(0),,-0.0152,-0.1106,-0.1766,-0.1766
+"Density index up to 10, if 1+ workers","@(df.workers>0)*df.density_index.clip(0, 10)",,0,-0.2028,-0.3654,-0.3654
+"Density index in excess of 10, if 1+ workers",@(df.workers>0)*(df.density_index-10).clip(0),,-0.0152,-0.1106,-0.1766,-0.1766
+Constants,@1,,1.1865,-1.0846,-3.2502,-5.313
+San Francisco county,county_name == 'San Francisco',,0.4259,0.4683,0.1458,0.1458
+Solano county,county_name == 'Solano',,-0.566,-0.4429,-0.2372,-0.2372
+Napa county,county_name == 'Napa',,-0.566,-0.4429,-0.2372,-0.2372
+Sonoma county,county_name == 'Sonoma',,-0.566,-0.4429,-0.2372,-0.2372
+Marin county,county_name == 'Marin',,-0.2434,0,0,0
+"Retail accessibility (0.66*PK + 0.34*OP) by auto, if 0 workers",(workers==0)*(0.66*AUTOPEAKRETAIL+0.34*AUTOOFFPEAKRETAIL),,0.0626,0.0626,0.0626,0.0626
+"Retail accessibility (0.66*PK + 0.34*OP) by auto, if 1+ workers",(workers>0)*(0.66*AUTOPEAKRETAIL+0.34*AUTOOFFPEAKRETAIL),,0.1646,0.1646,0.1646,0.1646
+"Retail accessibility (0.66*PK + 0.34*OP) by transit, if 0 workers",(workers==0)*(0.66*TRANSITPEAKRETAIL+0.34*TRANSITOFFPEAKRETAIL),,-0.3053,-0.3053,-0.3053,-0.3053
+"Retail accessibility (0.66*PK + 0.34*OP) by transit, if 1+ workers",(workers>0)*(0.66*TRANSITPEAKRETAIL+0.34*TRANSITOFFPEAKRETAIL),,-0.5117,-0.5117,-0.5117,-0.5117
+"Retail accessibility by non-motorized, if 0 workers",(workers==0)*NONMOTORIZEDRETAIL,,-0.03,-0.03,-0.03,-0.03
+"Retail accessibility by non-motorized, if 1+ workers",(workers>0)*NONMOTORIZEDRETAIL,,-0.03,-0.03,-0.03,-0.03
+"Auto time savings per worker (over walk or transit, max 120) to work",workTourAutoTimeSavings/workers,,0.4707,0.6142,0.5705,0.7693

example/configs/mandatory_tour_frequency.csv

@@ -0,0 +1,100 @@
+Description,Expression,work1,work2,school1,school2,work_and_school
+Full-time worker alternative-specific constants,ptype == 1,0,-3.3781,,,
+Part-time worker alternative-specific constants,ptype == 2,0,-3.0476,,,
+University student alternative-specific constants,ptype == 3,2.166,-1.3965,0,-3.7429,0.1073
+Non-working adult alternative-specific constants,ptype == 4,,,,,
+Retired alternative-specific constants,ptype == 5,,,,,
+Driving-age child alternative-specific constants,ptype == 6,,,0,-3.136,-4.4362
+Pre-driving age child who is in school alternative-specific constants,ptype == 7,,,0,-3.9703,
+Female - Full-time worker interaction,(ptype == 1) & (sex == 2),0,-0.2255,0.1592,,-0.3442
+Female - Part-time worker interaction,(ptype == 2) & (sex == 2),0,-0.2255,0.1592,,-0.3442
+Female - University student interaction,(ptype == 3) & (sex == 2),0.1737,-0.2255,0.1592,0.114,-0.3442
+Female - Non-working adult interaction,(ptype == 4) & (sex == 2),0,-0.2255,0.1592,,
+Female - Retired interaction,(ptype == 5) & (sex == 2),0,-0.2255,0.1592,,
+Female - Driving-age child interaction,(ptype == 6) & (sex == 2),0.1737,,0,0.114,-0.3442
+Female - Pre-driving age child who is in school interaction,(ptype == 7) & (sex == 2),0.1737,,0,0.114,
+Under 35 - Full-time worker interaction,(ptype == 1) & (age <= 35),0,-0.1375,0.7218,,0.9761
+Under 35 - Part-time worker interaction,(ptype == 2) & (age <= 35),0,-0.1375,0.7218,,0.9761
+Under 35 - University student interaction,(ptype == 3) & (age <= 35),-0.4629,-0.1375,0,1.275,0.9761
+Under 35 - Non-working adult interaction,(ptype == 4) & (age <= 35),0,-0.1375,0.7218,,
+Can walk to work - Full-time worker interaction,(ptype == 1) & (distance_to_work < 3),,0.5268,,,
+Can walk to work - Part-time worker interaction,(ptype == 2) & (distance_to_work < 3),,0.5268,,,
+Can walk to work - University student interaction,(ptype == 3) & (distance_to_work < 3),,0.5268,,,
+Can walk to work - Non-working adult interaction,(ptype == 4) & (distance_to_work < 3),,0.5268,,,
+Can walk to work - Retired interaction,(ptype == 5) & (distance_to_work < 3),,0.5268,,,
+Can walk to school - University student interaction,(ptype == 3) & (distance_to_school < 3),,,,0.7114,
+Can walk to school - Driving-age child interaction,(ptype == 6) & (distance_to_school < 3),,,,0.7114,
+Can walk to school - Pre-driving age child who is in school interaction,(ptype == 7) & (distance_to_school < 3),,,,0.7114,
+Can walk to work or school - Full-time worker interaction,(ptype == 1) & (distance_to_work < 3 | distance_to_school < 3),,,,,0.1391
+Can walk to work or school - Part-time worker interaction,(ptype == 2) & (distance_to_work < 3 | distance_to_school < 3),,,,,0.1391
+Can walk to work or school - University student interaction,(ptype == 3) & (distance_to_work < 3 | distance_to_school < 3),,,,,0.1391
+Can walk to work or school - Driving-age child interaction,(ptype == 6) & (distance_to_work < 3 | distance_to_school < 3),,,,,0.1391
+Round trip auto time to work - Full-time worker interaction,(ptype == 1) * roundtrip_auto_time_to_work,,-0.0035,,,-0.0031
+Round trip auto time to work - Part-time worker interaction,(ptype == 2) * roundtrip_auto_time_to_work,,-0.0035,,,-0.0031
+Round trip auto time to work - University student interaction,(ptype == 3) * roundtrip_auto_time_to_work,,-0.0035,,,-0.0031
+Round trip auto time to work - Non-working adult interaction,(ptype == 4) * roundtrip_auto_time_to_work,,-0.0035,,,
+Round trip auto time to work - Retired,(ptype == 5) * roundtrip_auto_time_to_work,,-0.0035,,,
+Round trip auto time to school - University student interaction,(ptype == 3) * roundtrip_auto_time_to_school,,,,-0.0034,-0.0031
+Round trip auto time to school - Driving-age child interaction,(ptype == 6) * roundtrip_auto_time_to_school,,,,-0.0034,-0.0031
+Round trip auto time to school - Pre-driving age child who is in school interaction,(ptype == 7) * roundtrip_auto_time_to_school,,,,-0.0034,
+Student is employed - University student interaction,(ptype == 3) & student_is_employed,3.014,3.014,,,3.014
+Student is employed - Driving-age child interaction,(ptype == 6) & student_is_employed,3.014,3.014,,,3.014
+Non-student goes to school - Full-time worker interaction,(ptype == 1) & nonstudent_to_school,,,3.883,,3.883
+Non-student goes to school - Part-time worker interaction,(ptype == 2) & nonstudent_to_school,,,3.883,,3.883
+Non-student goes to school - Non-working adult interaction,(ptype == 4) & nonstudent_to_school,,,3.883,,
+Non-student goes to school - Retired interaction,(ptype == 5) & nonstudent_to_school,,,3.883,,
+No cars in household - Full-time worker interaction,(ptype == 1) & (auto_ownership == 0),,-1.306,,,-1.302
+No cars in household - Part-time worker interaction,(ptype == 2) & (auto_ownership == 0),,-1.306,,,-1.302
+No cars in household - University student interaction,(ptype == 3) & (auto_ownership == 0),,-1.306,,-1.413,-1.302
+No cars in household - Non-working adult interaction,(ptype == 4) & (auto_ownership == 0),,-1.306,,,
+No cars in household - Retired interaction,(ptype == 5) & (auto_ownership == 0),,-1.306,,,
+No cars in household - Driving-age student interaction,(ptype == 6) & (auto_ownership == 0),,,,-1.413,-1.302
+No cars in household - Pre-driving age child who is in school interaction,(ptype == 7) & (auto_ownership == 0),,,,-1.413,
+Fewer cars than drivers in household - University student interaction,(ptype == 3) & (auto_ownership < drivers),,,,-0.5759,
+Fewer cars than drivers in household - Driving-age student interaction,(ptype == 6) & (auto_ownership < drivers),,,,-0.5759,
+Fewer cars than drivers in household - Pre-driving age child who is in school interaction,(ptype == 7) & (auto_ownership < drivers),,,,-0.5759,
+Number of preschool children in household - Full-time worker interaction,(ptype == 1) * (num_young_children),0,-0.1478,-0.1335,,-0.1251
+Number of preschool children in household - Part-time worker interaction,(ptype == 2) * (num_young_children),0,-0.1478,-0.1335,,-0.1251
+Number of preschool children in household - University student interaction,(ptype == 3) * (num_young_children),0.2191,-0.1478,0,-0.5577,-0.1251
+Number of preschool children in household - Non-working adult interaction,(ptype == 4) * (num_young_children),0,-0.1478,-0.1335,,
+Number of preschool children in household - Retired interaction,(ptype == 5) * (num_young_children),0,-0.1478,-0.1335,,
+Number of preschool children in household - Driving-age student interaction,(ptype == 6) * (num_young_children),0.2191,,0,-0.5577,-0.1251
+Number of preschool children in household - Pre-driving age child who is in school interaction,(ptype == 7) * (num_young_children),0.2191,,0,-0.5577,
+Number of non-workers in the household - Full-time worker interaction,(ptype == 1) * non_workers,,,0.2574,,
+Number of non-workers in the household - Part-time worker interaction,(ptype == 2) * non_workers,,,0.2574,,
+Household income higher than $50k - Full-time worker interaction,(ptype == 1) & (income_in_thousands > 50),0,,0.0347,,0.0347
+Household income higher than $50k - Part-time worker interaction,(ptype == 2) & (income_in_thousands > 50),0,,0.0347,,0.0347
+Household income higher than $50k - University student interaction,(ptype == 3) & (income_in_thousands > 50),-0.0528,-0.0528,0,,-0.0528
+Household income higher than $50k - Non-working adult interaction,(ptype == 4) & (income_in_thousands > 50),0,,0.0347,,
+Household income higher than $50k - Retired interaction,(ptype == 5) & (income_in_thousands > 50),0,,0.0347,,
+Household income higher than $50k - Driving-age student interaction,(ptype == 6) & (income_in_thousands > 50),-0.0528,,0,,-0.0528
+Household income higher than $50k - Pre-driving age child who is in school interaction,(ptype == 7) & (income_in_thousands > 50),-0.0528,,0,,
+Non-family household - Full-time worker interaction,(ptype == 1) & non_family,0,,-0.25,,-0.25
+Non-family household - Part-time worker interaction,(ptype == 2) & non_family,0,,-0.25,,-0.25
+Non-family household - University student interaction,(ptype == 3) & non_family,-0.1792,-0.1792,0,,-0.1792
+Non-family household - Non-working adult interaction,(ptype == 4) & non_family,0,,-0.25,,
+Non-family household - Retired interaction,(ptype == 5) & non_family,0,,-0.25,,
+Non-family household - Driving-age student interaction,(ptype == 6) & non_family,-0.1792,,0,,-0.1792
+Non-family household - Pre-driving age child who is in school interaction,(ptype == 7) & non_family,-0.1792,,0,,
+Number of children under 16 not at school - Full-time worker interaction,(ptype == 1) * num_under16_not_at_school,,0.1804,,,-0.1955
+Number of children under 16 not at school - Part-time worker interaction,(ptype == 2) * num_under16_not_at_school,,0.1804,,,-0.1955
+Number of children under 16 not at school - University student interaction,(ptype == 3) * num_under16_not_at_school,,0.1804,,0.0866,-0.1955
+Number of children under 16 not at school - Non-working adult interaction,(ptype == 4) * num_under16_not_at_school,,0.1804,,,
+Number of children under 16 not at school - Retired,(ptype == 5) * num_under16_not_at_school,,0.1804,,,
+Number of children under 16 not at school - Driving-age student interaction,(ptype == 6) * num_under16_not_at_school,,,,0.0866,-0.1955
+Number of children under 16 not at school - Pre-driving age child who is in school interaction,(ptype == 7) * num_under16_not_at_school,,,,0.0866,
+Home is in urban area - Full-time worker interaction,(ptype == 1) & (area_type < 4),0,0.2308,-0.1361,,-0.3509
+Home is in urban area - Part-time worker interaction,(ptype == 2) & (area_type < 4),0,0.2308,-0.1361,,-0.3509
+Home is in urban area - University student interaction,(ptype == 3) & (area_type < 4),-0.2831,0.2308,0,0.317,-0.3509
+Home is in urban area - Non-working adult interaction,(ptype == 4) & (area_type < 4),0,0.238,-0.1361,,
+Home is in urban area - Retired interaction,(ptype == 5) & (area_type < 4),0,0.2308,-0.1361,,
+Home is in urban area - Driving-age student interaction,(ptype == 6) & (area_type < 4),-0.2831,,0,0.317,-0.3509
+Home is in urban area - Pre-driving age child who is in school interaction,(ptype == 7) & (area_type < 4),-0.2831,,0,0.317,
+Unavailable: Full-time worker,ptype == 1,,,,-999,
+Unavailable: Part-time worker,ptype == 2,,,,-999,
+Unavailable: Non-working adult,ptype == 4,,,,-999,-999
+Unavailable: Retired,ptype == 5,,,,-999,-999
+Unavailable: Driving-age child,ptype == 6,-999,-999,,,
+Unavailable: Pre-driving age child who is in school,ptype == 7,,-999,,,-999
+Unavailable: Work tours for those with no usual work location,~(workplace_taz > -1),-999,-999,,,-999
+Unavailalbe: School tours for those with no usual school location,~(school_taz > -1),,,-999,-999,-999

example/configs/settings.yaml

@@ -1,6 +1,6 @@
 store: mtc_asim.h5
 
-households_sample_size: 100000
+households_sample_size: 10000
 
 county_map:
  San Francisco: 1
@@ -12,3 +12,34 @@ county_map:
  Napa: 7
  Sonoma: 8
  Marin: 9
+
+employment_map:
+ 1: "full"
+ 2: "part"
+ 3: "not"
+ 4: "child"
+
+student_map:
+ 1: "high"
+ 2: "college"
+ 3: "not"
+
+person_type_map:
+ 1: "full"
+ 2: "part"
+ 3: "university"
+ 4: "nonwork"
+ 5: "retired"
+ 6: "driving"
+ 7: "school"
+ 8: "preschool"
+
+household_type_map:
+ 0: "null"
+ 1: "family_married"
+ 2: "family_male"
+ 3: "family_female"
+ 4: "nonfamily_male_alone"
+ 5: "nonfamily_male_notalone"
+ 6: "nonfamily_female_alone"
+ 7: "nonfamily_female_notalone"