diff --git a/compile.py b/compile.py
index 4707f33..81df742 100644
--- a/compile.py
+++ b/compile.py
@@ -67,11 +67,11 @@ def run_notebooks(
nb_in = nbparam.replace_definitions(nb_in, new_params)
# execute it
- ep = ExecutePreprocessor(timeout=600, kernel_name="python3")
+ ep = ExecutePreprocessor(timeout=1000, kernel_name="python3")
ep.preprocess(nb_in)
# strip out all metadata that causes issues for Quarto
- cmp = ClearMetadataPreprocessor(timeout=600, kernel_name="python3")
+ cmp = ClearMetadataPreprocessor(timeout=1000, kernel_name="python3")
# exclude cell tags from removal
cmp.preserve_cell_metadata_mask |= {"tags"}
cmp.preprocess(nb_in, resources={})
diff --git a/notebooks/daily.ipynb b/notebooks/daily.ipynb
index 0258718..541a526 100644
--- a/notebooks/daily.ipynb
+++ b/notebooks/daily.ipynb
@@ -139,10 +139,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by home MAZ. Original Persons: {len(base_persons_df)}\")\n",
+ "print(f\"Filtering all persons by home MAZ. Original persons: {len(base_persons_df)}\")\n",
"base_persons_df = base_persons_df[single_filter_mazs(base_persons_df.home_zone_id)]\n",
"build_persons_df = build_persons_df[single_filter_mazs(build_persons_df.home_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_persons_df)}\")"
+ "print(f\"Total persons after filtering: {len(base_persons_df)}\")"
]
},
{
@@ -218,10 +218,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by home MAZ. Original Persons: {len(base_persons_df)}\")\n",
+ "print(f\"Filtering workers and students by home MAZ. Original workers/students: {len(base_persons_df)}\")\n",
"base_persons_df = base_persons_df[single_filter_mazs(base_persons_df.home_zone_id)]\n",
"build_persons_df = build_persons_df[single_filter_mazs(build_persons_df.home_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_persons_df)}\")"
+ "print(f\"Total workers/students after filtering: {len(base_persons_df)}\")"
]
},
{
@@ -334,10 +334,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by home MAZ. Original Persons: {len(base_persons_df)}\")\n",
+ "print(f\"Filtering all persons by home MAZ. Original persons: {len(base_persons_df)}\")\n",
"base_persons_df = base_persons_df[single_filter_mazs(base_persons_df.home_zone_id)]\n",
"build_persons_df = build_persons_df[single_filter_mazs(build_persons_df.home_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_persons_df)}\")"
+ "print(f\"Total persons after filtering: {len(base_persons_df)}\")"
]
},
{
diff --git a/notebooks/joint.ipynb b/notebooks/joint.ipynb
index 60eb573..da8f7cd 100644
--- a/notebooks/joint.ipynb
+++ b/notebooks/joint.ipynb
@@ -124,9 +124,9 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"\"\"Filtering tours by origin {\n",
+ "print(f\"\"\"Filtering joint tours by origin {\n",
" 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
- " } destination MAZ.\\nOriginal tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\"\"\")\n",
+ " } destination MAZ.\\nOriginal joint tours in base: {len(base_tour[base_tour.number_of_participants > 1])}\\tbuild: {len(build_tour[build_tour.number_of_participants > 1])}\"\"\")\n",
"\n",
"base_tour = base_tour[multi_filter_mazs([base_tour.origin, base_tour.destination])] # base tour in the filtered set\n",
"\n",
@@ -139,7 +139,7 @@
" & multi_filter_mazs([build_tour.origin, build_tour.destination]) # and it's in this set\n",
" )]\n",
"\n",
- "print(f\"After filtering, tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\")"
+ "print(f\"After filtering, joint tours in base: {len(base_tour[base_tour.number_of_participants > 1])}\\tbuild: {len(build_tour[build_tour.number_of_participants > 1])}\")"
]
},
{
@@ -389,7 +389,7 @@
" left_index=True,\n",
" right_index=True,\n",
" suffixes=('_base', '_build'),\n",
- " how='inner'\n",
+ " how='outer'\n",
")\n",
"\n",
"# get the difference in start and end times\n",
@@ -448,18 +448,19 @@
" 3: \"PM\",\n",
" 4: \"EV\"\n",
"}\n",
- "tp_order = ['EA','AM','MD','PM','EV','Total']\n",
+ "tp_order_base = ['EA','AM','MD','PM','EV','Newly created','Total']\n",
+ "tp_order_build = ['EA','AM','MD','PM','EV','Removed','Total']\n",
"\n",
"for metric in ['start', 'end']:\n",
" purpose_df = df\n",
" display(Markdown(f\"### Joint tour {metric} changes\"))\n",
" xtab = pd.crosstab(\n",
- " purpose_df[f'{metric}_period_base'].replace(mapper),\n",
- " purpose_df[f'{metric}_period_build'].replace(mapper),\n",
+ " purpose_df[f'{metric}_period_base'].replace(mapper).fillna('Newly created'),\n",
+ " purpose_df[f'{metric}_period_build'].replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
" )\n",
- " display(xtab.loc[tp_order,tp_order])\n",
+ " display(xtab.loc[tp_order_base,tp_order_build])\n",
"\n",
" display(Markdown(\" \"))\n",
"\n"
diff --git a/notebooks/long_term.ipynb b/notebooks/long_term.ipynb
index ac3eda1..6290c00 100644
--- a/notebooks/long_term.ipynb
+++ b/notebooks/long_term.ipynb
@@ -25,6 +25,7 @@
"import pandas as pd\n",
"import yaml\n",
"import plotly.io as pio\n",
+ "import numpy as np\n",
"pio.renderers.default = \"plotly_mimetype+notebook_connected\"\n",
"\n",
"from IPython.display import Markdown, display\n",
@@ -185,10 +186,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by home MAZ. Original Persons: {len(base_per)}\")\n",
+ "print(f\"Filtering workers by home MAZ. Original workers: {len(base_per[base_per.is_worker])}\")\n",
"base_per = base_per[single_filter_mazs(base_per.home_zone_id)]\n",
"build_per = build_per[single_filter_mazs(build_per.home_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_per)}\")"
+ "print(f\"Workers after filtering: {len(base_per[base_per.is_worker])}\")"
]
},
{
@@ -238,10 +239,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by home MAZ. Original Persons: {len(base_per)}\")\n",
+ "print(f\"Filtering workers by home MAZ. Original workers: {len(base_per[base_per.is_worker])}\")\n",
"base_per = base_per[single_filter_mazs(base_per.home_zone_id)]\n",
"build_per = build_per[single_filter_mazs(build_per.home_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_per)}\")"
+ "print(f\"Workers after filtering: {len(base_per[base_per.is_worker])}\")"
]
},
{
@@ -303,10 +304,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by home MAZ. Original Persons: {len(base_per)}\")\n",
+ "print(f\"Filtering students by home MAZ. Original students: {len(base_per[base_per.is_student])}\")\n",
"base_per = base_per[single_filter_mazs(base_per.home_zone_id)]\n",
"build_per = build_per[single_filter_mazs(build_per.home_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_per)}\")"
+ "print(f\"Students after filtering: {len(base_per[base_per.is_student])}\")"
]
},
{
@@ -370,10 +371,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by home MAZ. Original Persons: {len(base_per)}\")\n",
+ "print(f\"Filtering all persons by home MAZ. Original persons: {len(base_per)}\")\n",
"base_per = base_per[single_filter_mazs(base_per.home_zone_id)]\n",
"build_per = build_per[single_filter_mazs(build_per.home_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_per)}\")"
+ "print(f\"Total persons after filtering: {len(base_per)}\")"
]
},
{
@@ -531,10 +532,10 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by workplace MAZ. Original Persons: {len(base_per)}\")\n",
+ "print(f\"Filtering workers by workplace MAZ. Original workers in base scenario: {len(base_per[base_per.is_worker])}\")\n",
"base_per = base_per[single_filter_mazs(base_per.workplace_zone_id)]\n",
"build_per = build_per[single_filter_mazs(build_per.workplace_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_per)}\")"
+ "print(f\"Workers in base scenario after filtering: {len(base_per[base_per.is_worker])}\")"
]
},
{
@@ -545,7 +546,7 @@
"outputs": [],
"source": [
"df = base_per[base_per.is_worker].merge(build_per[build_per.is_worker],\n",
- " how='left',\n",
+ " how='outer',\n",
" left_index=True,\n",
" right_index=True, \n",
" suffixes=('_base','_build'))"
@@ -559,12 +560,13 @@
"outputs": [],
"source": [
"df0 = pd.crosstab(\n",
- " base_per[base_per.is_worker].free_parking_at_work,\n",
- " build_per[build_per.is_worker].free_parking_at_work,\n",
+ " df.free_parking_at_work_base.map({False:\"No free parking\",True:\"Has free parking\",np.nan:\"Workplace moved into area\"}),\n",
+ " df.free_parking_at_work_build.map({False:\"No free parking\",True:\"Has free parking\",np.nan:\"Workplace moved out of area\"}),\n",
" rownames=['base'],\n",
" colnames=['build'],\n",
" margins=True,\n",
- " margins_name='Total'\n",
+ " margins_name='Total',\n",
+ " dropna=False,\n",
")\n",
"df0"
]
@@ -576,9 +578,13 @@
"metadata": {},
"outputs": [],
"source": [
- "df = (df.free_parking_at_work_base == df.free_parking_at_work_build).value_counts()\n",
- "df.index = df.index.map({True:'Unchanged',False:'Changed'})\n",
- "df"
+ "df.loc[df.free_parking_at_work_base == df.free_parking_at_work_build,\"Case\"] = \"Unchanged\"\n",
+ "df.loc[df.free_parking_at_work_base != df.free_parking_at_work_build,\"Case\"] = \"Changed\"\n",
+ "\n",
+ "df.loc[df.free_parking_at_work_base.isna(),\"Case\"] = \"Workplace moved into area\"\n",
+ "df.loc[df.free_parking_at_work_build.isna(),\"Case\"] = \"Workplace moved out of area\"\n",
+ "\n",
+ "df.Case.value_counts().sort_index()"
]
},
{
@@ -588,7 +594,7 @@
"metadata": {},
"outputs": [],
"source": [
- "fig = vh.create_pie_chart(df.to_frame().sort_index(),[\"count\"])\n",
+ "fig = vh.create_pie_chart(df.Case.value_counts().to_frame().sort_index(),[\"count\"])\n",
"fig.show()"
]
},
@@ -607,7 +613,7 @@
"metadata": {},
"outputs": [],
"source": [
- "usecols = ['person_id', 'is_worker', 'telecommute_frequency', 'workplace_zone_id']\n",
+ "usecols = ['person_id', 'is_worker', 'telecommute_frequency', 'workplace_zone_id', 'work_from_home']\n",
"base_per = pd.read_csv(f\"{base_dir}/final_persons.csv\",\n",
" index_col='person_id',\n",
" usecols=usecols)\n",
@@ -635,10 +641,12 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"Filtering persons by workplace MAZ. Original Persons: {len(base_per)}\")\n",
+ "base_tele_mask = (base_per.is_worker) & ~(base_per.work_from_home)\n",
+ "build_tele_mask = (build_per.is_worker) & ~(build_per.work_from_home)\n",
+ "print(f\"Filtering workers by workplace MAZ who do not work from home. Original workers in base scenario: {len(base_per[base_tele_mask])}\")\n",
"base_per = base_per[single_filter_mazs(base_per.workplace_zone_id)]\n",
"build_per = build_per[single_filter_mazs(build_per.workplace_zone_id)]\n",
- "print(f\"Persons after filtering: {len(base_per)}\")"
+ "print(f\"Workers in base scenario after filtering: {len(base_per[base_tele_mask])}\")"
]
},
{
@@ -648,8 +656,8 @@
"metadata": {},
"outputs": [],
"source": [
- "df = base_per[base_per.is_worker].merge(build_per[build_per.is_worker],\n",
- " how='left',\n",
+ "df = base_per[base_tele_mask].merge(build_per[build_tele_mask],\n",
+ " how='outer',\n",
" left_index=True,\n",
" right_index=True, \n",
" suffixes=('_base','_build'))"
@@ -663,14 +671,18 @@
"outputs": [],
"source": [
"xtab = pd.crosstab(\n",
- " df.telecommute_frequency_base,\n",
- " df.telecommute_frequency_build,\n",
+ " df.telecommute_frequency_base.fillna(\"Workplace moved into area\"),\n",
+ " df.telecommute_frequency_build.fillna(\"Workplace moved out of area\"),\n",
" rownames=['base'],\n",
" colnames=['build'],\n",
" margins=True,\n",
- " margins_name='Total'\n",
- ").sort_index()\n",
- "xtab"
+ " margins_name='Total',\n",
+ " dropna=False,\n",
+ ")\n",
+ "xtab.loc[\n",
+ " filter(lambda x: x in xtab.index, ['0 (No Telecommute)','1 Day per Week','2-3 Days per Week','4 Days per Week','Workplace moved into area','Total']),\n",
+ " filter(lambda x: x in xtab.columns, ['0 (No Telecommute)','1 Day per Week','2-3 Days per Week','4 Days per Week','Workplace moved out of area','Total'])\n",
+ " ]"
]
},
{
@@ -688,6 +700,8 @@
"df1.loc[base_tc_magnitude == build_tc_magnitude, 'case'] = 'Unchanged'\n",
"df1.loc[base_tc_magnitude > build_tc_magnitude, 'case'] = 'Decreased'\n",
"df1.loc[base_tc_magnitude < build_tc_magnitude, 'case'] = 'Increased'\n",
+ "df1.loc[df1.telecommute_frequency_base.isna() & (~df1.telecommute_frequency_build.isna()),'case'] = \"Workplace moved into area\"\n",
+ "df1.loc[df1.telecommute_frequency_build.isna() & (~df1.telecommute_frequency_base.isna()),'case'] = \"Workplace moved out of area\"\n",
"\n",
"fig = vh.create_pie_chart(df1.case.value_counts().to_frame().sort_index(), [\"count\"])\n",
"fig.show()"
@@ -764,7 +778,7 @@
],
"metadata": {
"kernelspec": {
- "display_name": "asimviz",
+ "display_name": "asim_eet_viz",
"language": "python",
"name": "python3"
},
diff --git a/notebooks/tours.ipynb b/notebooks/tours.ipynb
index 56855ce..1324dfb 100644
--- a/notebooks/tours.ipynb
+++ b/notebooks/tours.ipynb
@@ -88,15 +88,16 @@
"metadata": {},
"outputs": [],
"source": [
+ "cols_to_use = ['tour_id','tour_category','origin','destination','primary_purpose','person_id','start','end','tour_type','tour_mode','parent_tour_id','stop_frequency']\n",
"base_tour = pd.read_csv(f\"{base_dir}/final_tours.csv\", \n",
" index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','origin','destination','primary_purpose','person_id','start','end','tour_type'])\n",
+ " usecols=cols_to_use)\n",
"\n",
"base_tour_idx = base_tour.index\n",
"\n",
"build_tour = pd.read_csv(f\"{build_dir}/final_tours.csv\", \n",
" index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','origin','destination','primary_purpose','person_id','start','end','tour_type'])\n"
+ " usecols=cols_to_use)\n"
]
},
{
@@ -106,7 +107,7 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"\"\"Filtering tours by origin {\n",
+ "print(f\"\"\"Filtering all tours by origin {\n",
" 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
" } destination MAZ.\\nOriginal tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\\tdiff: {len(build_tour)-len(base_tour)}\"\"\")\n",
"\n",
@@ -121,7 +122,7 @@
" & multi_filter_mazs([build_tour.origin, build_tour.destination]) # and it's in this set\n",
" )]\n",
"\n",
- "print(f\"After filtering, tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\\tdiff: {len(build_tour)-len(base_tour)}\")"
+ "print(f\"After filtering, total tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\\tdiff: {len(build_tour)-len(base_tour)}\")"
]
},
{
@@ -284,8 +285,6 @@
"source": [
"# keep only work and school tours\n",
"base_tours_df = base_tour[base_tour.tour_type.isin(['work', 'school'])]\n",
- "base_tour_idx = base_tours_df.index\n",
- "\n",
"build_tours_df = build_tour[build_tour.tour_type.isin(['work', 'school'])]"
]
},
@@ -319,7 +318,7 @@
" left_index=True,\n",
" right_index=True,\n",
" suffixes=('_base', '_build'),\n",
- " how='inner'\n",
+ " how='outer'\n",
")\n",
"\n",
"# get the difference in start and end times\n",
@@ -372,14 +371,16 @@
" 3: \"PM\",\n",
" 4: \"EV\"\n",
"}\n",
- "tp_order = ['EA','AM','MD','PM','EV','Total']\n",
+ "tp_order_base = ['EA','AM','MD','PM','EV','Newly created','Total']\n",
+ "tp_order_build = ['EA','AM','MD','PM','EV','Removed','Total']\n",
+ "\n",
"xtab = pd.crosstab(\n",
- " df.start_period_base.replace(mapper),\n",
- " df.start_period_build.replace(mapper),\n",
+ " df.start_period_base.replace(mapper).fillna('Newly created'),\n",
+ " df.start_period_build.replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
")\n",
- "display(xtab.loc[tp_order,tp_order])"
+ "display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))"
]
},
{
@@ -402,16 +403,16 @@
"\n",
"for purpose in ['work', 'school']:\n",
" for metric in ['start']:\n",
- " purpose_df = df[df.tour_type_base == purpose]\n",
+ " purpose_df = df[(df.tour_type_base == purpose) | (df.tour_type_build == purpose)]\n",
" display(Markdown(f\"### {purpose.capitalize()}\"))\n",
"\n",
" xtab = pd.crosstab(\n",
- " purpose_df[f'{metric}_period_base'].replace(mapper),\n",
- " purpose_df[f'{metric}_period_build'].replace(mapper),\n",
+ " purpose_df[f'{metric}_period_base'].replace(mapper).fillna('Newly created'),\n",
+ " purpose_df[f'{metric}_period_build'].replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
" )\n",
- " display(xtab.loc[tp_order,tp_order])\n",
+ " display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))\n",
"\n",
"\n",
" summary_text = f\"\"\"**{purpose.capitalize()} tour {metric} changes:**\n",
@@ -457,12 +458,12 @@
"outputs": [],
"source": [
"xtab = pd.crosstab(\n",
- " df.end_period_base.replace(mapper),\n",
- " df.end_period_build.replace(mapper),\n",
+ " df.end_period_base.replace(mapper).fillna('Newly created'),\n",
+ " df.end_period_build.replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
")\n",
- "display(xtab.loc[tp_order,tp_order])"
+ "display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))"
]
},
{
@@ -485,16 +486,16 @@
"\n",
"for purpose in ['work', 'school']:\n",
" for metric in ['end']:\n",
- " purpose_df = df[df.tour_type_base == purpose]\n",
+ " purpose_df = df[(df.tour_type_base == purpose) | (df.tour_type_build == purpose)]\n",
" display(Markdown(f\"### {purpose.capitalize()}\"))\n",
"\n",
" xtab = pd.crosstab(\n",
- " purpose_df[f'{metric}_period_base'].replace(mapper),\n",
- " purpose_df[f'{metric}_period_build'].replace(mapper),\n",
+ " purpose_df[f'{metric}_period_base'].replace(mapper).fillna('Newly created'),\n",
+ " purpose_df[f'{metric}_period_build'].replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
" )\n",
- " display(xtab.loc[tp_order,tp_order])\n",
+ " display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))\n",
"\n",
"\n",
" summary_text = f\"\"\"**{purpose.capitalize()} tour {metric} changes:**\n",
@@ -574,7 +575,7 @@
" left_index=True,\n",
" right_index=True,\n",
" suffixes=('_base', '_build'),\n",
- " how='inner'\n",
+ " how='outer'\n",
")\n",
"\n",
"# get the difference in start and end times\n",
@@ -621,12 +622,12 @@
"outputs": [],
"source": [
"xtab = pd.crosstab(\n",
- " df.start_period_base.replace(mapper),\n",
- " df.start_period_build.replace(mapper),\n",
+ " df.start_period_base.replace(mapper).fillna('Newly created'),\n",
+ " df.start_period_build.replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
")\n",
- "display(xtab)"
+ "display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))"
]
},
{
@@ -652,18 +653,18 @@
"# above comment is needed for Quarto to render subtabs correctly\n",
"metric='start'\n",
"for key, value in name_dict.items():\n",
- " purpose_df = df[df.primary_purpose_base == key]\n",
+ " purpose_df = df[(df.primary_purpose_base == key) | (df.primary_purpose_build == key)]\n",
" if len(purpose_df) == 0:\n",
" continue\n",
" display(Markdown(f\"### {key.capitalize()}\"))\n",
"\n",
" xtab = pd.crosstab(\n",
- " purpose_df[f'{metric}_period_base'].replace(mapper),\n",
- " purpose_df[f'{metric}_period_build'].replace(mapper),\n",
+ " purpose_df[f'{metric}_period_base'].replace(mapper).fillna('Newly created'),\n",
+ " purpose_df[f'{metric}_period_build'].replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
" )\n",
- " display(xtab)\n",
+ " display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))\n",
"\n",
"\n",
"\n",
@@ -705,12 +706,12 @@
"outputs": [],
"source": [
"xtab = pd.crosstab(\n",
- " df.end_period_base.replace(mapper),\n",
- " df.end_period_build.replace(mapper),\n",
+ " df.end_period_base.replace(mapper).fillna('Newly created'),\n",
+ " df.end_period_build.replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
")\n",
- "display(xtab)"
+ "display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))"
]
},
{
@@ -736,18 +737,18 @@
"# above comment is needed for Quarto to render subtabs correctly\n",
"metric='end'\n",
"for key, value in name_dict.items():\n",
- " purpose_df = df[df.primary_purpose_base == key]\n",
+ " purpose_df = df[(df.primary_purpose_base == key) | (df.primary_purpose_build == key)]\n",
" if len(purpose_df) == 0:\n",
" continue\n",
" display(Markdown(f\"### {key.capitalize()}\"))\n",
"\n",
" xtab = pd.crosstab(\n",
- " purpose_df[f'{metric}_period_base'].replace(mapper),\n",
- " purpose_df[f'{metric}_period_build'].replace(mapper),\n",
+ " purpose_df[f'{metric}_period_base'].replace(mapper).fillna('Newly created'),\n",
+ " purpose_df[f'{metric}_period_build'].replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
" )\n",
- " display(xtab)\n",
+ " display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))\n",
"\n",
" summary_text = f\"\"\"**{key.capitalize()} tour {metric} changes:**\n",
" Tours which arrived earlier in build: {len(purpose_df[purpose_df[f'{metric}_bin_difference'] < 0])}\n",
@@ -791,48 +792,6 @@
"# xtab base-vs build mode choice for NM tours by purpose"
]
},
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "5752ee4a-e49d-4842-876e-435937812ab5",
- "metadata": {},
- "outputs": [],
- "source": [
- "usecols = ['tour_id','tour_category','origin','destination','tour_mode', 'tour_category', 'primary_purpose']\n",
- "base_tour = pd.read_csv(f\"{base_dir}/final_tours.csv\", \n",
- " index_col='tour_id', \n",
- " usecols=usecols)\n",
- "\n",
- "build_tour = pd.read_csv(f\"{build_dir}/final_tours.csv\", \n",
- " index_col='tour_id',\n",
- " usecols=usecols)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "9f592846-8521-479e-9435-e30edd401a12",
- "metadata": {},
- "outputs": [],
- "source": [
- "print(f\"\"\"Filtering tours by origin {\n",
- " 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
- " } destination MAZ.\\nOriginal tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\"\"\")\n",
- "\n",
- "base_tour = base_tour[multi_filter_mazs([base_tour.origin, base_tour.destination])] # base tour in the filtered set\n",
- "\n",
- "# special build-case filtering\n",
- "build_tour = build_tour[(\n",
- " build_tour.index.isin(base_tour_idx) # originally existed in base\n",
- " & build_tour.index.isin(base_tour.index) # and was in the current set for the base\n",
- " ) | # OR\n",
- " ((~build_tour.index.isin(base_tour_idx)) # is a new tour \n",
- " & multi_filter_mazs([build_tour.origin, build_tour.destination]) # and it's in this set\n",
- " )]\n",
- "\n",
- "print(f\"After filtering, tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\")"
- ]
- },
{
"cell_type": "code",
"execution_count": null,
@@ -901,10 +860,14 @@
" \"TNC_SINGLE\": \"RIDESHARE\",\n",
"\n",
"}\n",
- "order = ['DRIVEALONE','SHARED2','SHARED3',\n",
+ "mode_order_base = ['DRIVEALONE','SHARED2','SHARED3',\n",
" 'WALK','BIKE','MICROMOBILITY','WALK_TRANSIT',\n",
" 'PNR_TRANSIT','KNR_TRANSIT','TNC_TRANSIT',\n",
- " 'RIDESHARE','SCH_BUS','Newly created','Removed',\"Total\",]\n"
+ " 'RIDESHARE','SCH_BUS','Newly created',\"Total\",]\n",
+ "mode_order_build = ['DRIVEALONE','SHARED2','SHARED3',\n",
+ " 'WALK','BIKE','MICROMOBILITY','WALK_TRANSIT',\n",
+ " 'PNR_TRANSIT','KNR_TRANSIT','TNC_TRANSIT',\n",
+ " 'RIDESHARE','SCH_BUS','Removed',\"Total\",]\n"
]
},
{
@@ -914,16 +877,13 @@
"metadata": {},
"outputs": [],
"source": [
- "# xtab = df[[\"tour_mode_base\", \"tour_mode_build\"]].replace(combiners).value_counts(dropna=False).unstack().fillna(0)\n",
- "# xtab.loc[order,order]\n",
- "order = pd.Series(order)\n",
"xtab = pd.crosstab(\n",
" df.tour_mode_base.replace(combiners).fillna('Newly created'),\n",
" df.tour_mode_build.replace(combiners).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total')\n",
"\n",
- "display(xtab.loc[order[order.isin(xtab.index)],order[order.isin(xtab.columns)]])"
+ "display(xtab.reindex(index=mode_order_base,columns=mode_order_build,fill_value=0))"
]
},
{
@@ -1052,7 +1012,7 @@
"# only trips that exist in the base run will be output\n",
"# looping thru name_dict will not find cases where the base value for purpose or category is nan in the outer join\n",
"for key, value in name_dict.items():\n",
- " df_purp = df.loc[(df.primary_purpose_base == key)]\n",
+ " df_purp = df.loc[(df.primary_purpose_base == key)|(df.primary_purpose_build == key)]\n",
" if len(df_purp) == 0:\n",
" continue \n",
" \n",
@@ -1074,6 +1034,142 @@
":::"
]
},
+ {
+ "cell_type": "markdown",
+ "id": "b24b7dfb",
+ "metadata": {},
+ "source": [
+ "## Stop Frequency"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "ad15ecd1",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "df = pd.merge(\n",
+ " base_tour,\n",
+ " build_tour,\n",
+ " how='outer',\n",
+ " left_index=True,\n",
+ " right_index=True,\n",
+ " suffixes=['_base', '_build']\n",
+ ")\n",
+ "\n",
+ "df.loc[df.stop_frequency_base == df.stop_frequency_build,'Case'] = \"Unchanged\"\n",
+ "df.loc[df.stop_frequency_base != df.stop_frequency_build,'Case'] = \"Changed\"\n",
+ "\n",
+ "df.loc[df.stop_frequency_base.isna(),'Case'] = \"Newly Created\"\n",
+ "df.loc[df.stop_frequency_build.isna(),\"Case\"] = \"Removed\"\n",
+ "\n",
+ "df.Case.value_counts()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "791905f1",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "total_stops_base = (df.stop_frequency_base.str[0].astype(float) \n",
+ " + df.stop_frequency_base.str[5].astype(float)\n",
+ " ).fillna(-np.inf).clip(upper=4)\n",
+ "\n",
+ "total_stops_build = (df.stop_frequency_build.str[0].astype(float) \n",
+ " + df.stop_frequency_build.str[5].astype(float)\n",
+ " ).fillna(-np.inf).clip(upper=4)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "e0766c89",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "fig = vh.create_pie_chart(df.Case.value_counts().to_frame().sort_index(), [\"count\"])\n",
+ "fig"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "359f52ef",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "diff = total_stops_build - total_stops_base\n",
+ "xtab = pd.crosstab(\n",
+ " total_stops_base.replace({4:\"4+\",-np.inf:\"Newly created\"}).astype(str).str.replace(\".0\",\"\"),\n",
+ " total_stops_build.clip(upper=4).replace({4:\"4+\",-np.inf:\"Removed\"}).astype(str).str.replace(\".0\",\"\"),\n",
+ " margins=True,\n",
+ " margins_name='Total')\n",
+ "\n",
+ "display(xtab.sort_index())"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "331e7142",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "summary_text = f\"\"\"**Stop frequency changes:**\n",
+ " Tours with fewer stops in build: {len(diff[diff < 0])}\n",
+ " Tours with the same stops in build: {len(diff[diff==0])}\n",
+ " Tours with more stops in build: {len(diff[diff > 0])}\n",
+ "\"\"\"\n",
+ "display(Markdown(summary_text.replace(\"\\n\",\"
\")))"
+ ]
+ },
+ {
+ "cell_type": "raw",
+ "id": "21679760",
+ "metadata": {},
+ "source": [
+ "::: {.panel-tabset}"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "c04587c3",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# | output: asis\n",
+ "# above comment is needed for Quarto to render subtabs correctly\n",
+ "\n",
+ "# only trips that exist in the base run will be output\n",
+ "# looping thru name_dict will not find cases where the base value for purpose or category is nan in the outer join\n",
+ "for key, value in name_dict.items():\n",
+ " df_purp = df.loc[(df.primary_purpose_base == key) | (df.primary_purpose_build == key)]\n",
+ " if len(df_purp) == 0:\n",
+ " continue \n",
+ " \n",
+ " print(f\"#### {value}\")\n",
+ " df_purp_cases = df_purp.Case.value_counts()\n",
+ " print(str(df_purp_cases).replace(\"\\n\",\"
\"))\n",
+ " \n",
+ " fig = vh.create_pie_chart(df_purp_cases.to_frame().sort_index(),[\"count\"])\n",
+ " \n",
+ " fig.show()\n",
+ " display(Markdown(\" \"))"
+ ]
+ },
+ {
+ "cell_type": "raw",
+ "id": "1a466645",
+ "metadata": {},
+ "source": [
+ ":::"
+ ]
+ },
{
"cell_type": "markdown",
"id": "c1a39b2b",
@@ -1095,11 +1191,11 @@
"\n",
"base_tour = pd.read_csv(f\"{base_dir}/final_tours.csv\", \n",
" index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','primary_purpose','destination', 'parent_tour_id'])\n",
+ " usecols=['tour_id','tour_category','primary_purpose','origin','destination', 'parent_tour_id'])\n",
"\n",
"build_tour = pd.read_csv(f\"{build_dir}/final_tours.csv\", \n",
" index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','primary_purpose','destination', 'parent_tour_id'])\n",
+ " usecols=['tour_id','tour_category','primary_purpose','origin','destination', 'parent_tour_id'])\n",
"\n",
"# keep work tours only\n",
"base_work_tour = base_tour[base_tour.primary_purpose == 'work']\n",
@@ -1114,9 +1210,11 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"\"\"Filtering work tours by destination MAZ.\\nOriginal tours in base: {len(base_work_tour)}\\tbuild: {len(build_work_tour)}\\tdiff: {len(build_work_tour)-len(base_work_tour)}\"\"\")\n",
+ "print(f\"\"\"Filtering work tours by origin {\n",
+ " 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
+ " } destination MAZ.\\nOriginal work tours in base: {len(base_work_tour)}\\tbuild: {len(build_work_tour)}\\tdiff: {len(build_work_tour)-len(base_work_tour)}\"\"\")\n",
"\n",
- "base_work_tour = base_work_tour[single_filter_mazs(base_work_tour.destination)] # base tour in the filtered set\n",
+ "base_work_tour = base_work_tour[multi_filter_mazs([base_work_tour.origin, base_work_tour.destination])] # base tour in the filtered set\n",
"\n",
"# special build-case filtering\n",
"build_work_tour = build_work_tour[(\n",
@@ -1124,10 +1222,10 @@
" & build_work_tour.index.isin(base_work_tour.index) # and was in the current set for the base\n",
" ) | # OR\n",
" ((~build_work_tour.index.isin(base_work_tour_idx)) # is a new tour \n",
- " & single_filter_mazs(build_work_tour.destination)# and it's in this set\n",
+ " & multi_filter_mazs([build_work_tour.origin, build_work_tour.destination]) # and it's in this set\n",
" )]\n",
"\n",
- "print(f\"After filtering, tours in base: {len(base_work_tour)}\\tbuild: {len(build_work_tour)}\\tdiff: {len(build_work_tour)-len(base_work_tour)}\")"
+ "print(f\"After filtering, work tours in base: {len(base_work_tour)}\\tbuild: {len(build_work_tour)}\\tdiff: {len(build_work_tour)-len(base_work_tour)}\")"
]
},
{
@@ -1192,11 +1290,11 @@
"source": [
"base_tour = pd.read_csv(f\"{base_dir}/final_tours.csv\", \n",
" index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','origin','destination','primary_purpose','start','end'])\n",
+ " usecols=['tour_id','tour_category','origin','destination','primary_purpose','start','end','tour_mode'])\n",
"\n",
"build_tour = pd.read_csv(f\"{build_dir}/final_tours.csv\", \n",
" index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','origin','destination','primary_purpose','start','end'])\n"
+ " usecols=['tour_id','tour_category','origin','destination','primary_purpose','start','end','tour_mode'])\n"
]
},
{
@@ -1206,9 +1304,16 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"\"\"Filtering tours by origin {\n",
+ "print(f\"\"\"Filtering at-work subtours by origin {\n",
" 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
- " } destination MAZ.\\nOriginal tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\\tdiff: {len(build_tour)-len(base_tour)}\"\"\")\n",
+ " } destination MAZ.\\nOriginal subtours in base: {\n",
+ " len(base_tour[base_tour.tour_category == 'atwork'])\n",
+ " }\\tbuild: {\n",
+ " len(build_tour[build_tour.tour_category == 'atwork'])\n",
+ " }\\tdiff: {\n",
+ " len(build_tour[build_tour.tour_category == 'atwork'])\n",
+ " -len(base_tour[base_tour.tour_category == 'atwork'])\n",
+ " }\"\"\")\n",
"\n",
"base_tour = base_tour[multi_filter_mazs([base_tour.origin, base_tour.destination])] # base tour in the filtered set\n",
"\n",
@@ -1221,7 +1326,11 @@
" & multi_filter_mazs([build_tour.origin, build_tour.destination]) # and it's in this set\n",
" )]\n",
"\n",
- "print(f\"After filtering, tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\")"
+ "print(f\"\"\"After filtering, at-work subtours in base: {\n",
+ " len(base_tour[base_tour.tour_category == 'atwork'])\n",
+ " }\\tbuild: {\n",
+ " len(build_tour[build_tour.tour_category == 'atwork'])\n",
+ " }\"\"\")"
]
},
{
@@ -1283,51 +1392,6 @@
"## At-Work Subtour Scheduling"
]
},
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "ee088c27",
- "metadata": {},
- "outputs": [],
- "source": [
- "base_tour = pd.read_csv(f\"{base_dir}/final_tours.csv\", \n",
- " index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','origin','destination','primary_purpose','start','end'])\n",
- "base_tour = base_tour[base_tour.tour_category == 'atwork']\n",
- "\n",
- "base_tour_idx = base_tour.index\n",
- "\n",
- "build_tour = pd.read_csv(f\"{build_dir}/final_tours.csv\", \n",
- " index_col='tour_id', \n",
- " usecols=['tour_id','tour_category','origin','destination','primary_purpose','start','end'])\n",
- "build_tour = build_tour[build_tour.tour_category == 'atwork']"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "e8ab8310",
- "metadata": {},
- "outputs": [],
- "source": [
- "print(f\"\"\"Filtering tours by origin {\n",
- " 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
- " } destination MAZ.\\nOriginal tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\\tdiff: {len(build_tour)-len(base_tour)}\"\"\")\n",
- "\n",
- "base_tour = base_tour[multi_filter_mazs([base_tour.origin, base_tour.destination])] # base tour in the filtered set\n",
- "\n",
- "# special build-case filtering\n",
- "build_tour = build_tour[(\n",
- " build_tour.index.isin(base_tour_idx) # originally existed in base\n",
- " & build_tour.index.isin(base_tour.index) # and was in the current set for the base\n",
- " ) | # OR\n",
- " ((~build_tour.index.isin(base_tour_idx)) # is a new tour \n",
- " & multi_filter_mazs([build_tour.origin, build_tour.destination]) # and it's in this set\n",
- " )]\n",
- "\n",
- "print(f\"After filtering, tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\")"
- ]
- },
{
"cell_type": "code",
"execution_count": null,
@@ -1356,12 +1420,12 @@
"# join base and build tours on tour_id\n",
"# compare tours that are in both base and build datasets\n",
"df = pd.merge(\n",
- " base_tour,\n",
- " build_tour,\n",
+ " base_tour[base_tour.tour_category == 'atwork'],\n",
+ " build_tour[build_tour.tour_category == 'atwork'],\n",
" left_index=True,\n",
" right_index=True,\n",
" suffixes=('_base', '_build'),\n",
- " how='inner'\n",
+ " how='outer'\n",
")\n",
"\n",
"# get the difference in start and end times\n",
@@ -1389,12 +1453,12 @@
"outputs": [],
"source": [
"xtab = pd.crosstab(\n",
- " df.start_period_base.replace(mapper),\n",
- " df.start_period_build.replace(mapper),\n",
+ " df.start_period_base.replace(mapper).fillna('Newly created'),\n",
+ " df.start_period_build.replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
")\n",
- "display(xtab.loc[tp_order,tp_order])"
+ "display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))"
]
},
{
@@ -1413,12 +1477,12 @@
"outputs": [],
"source": [
"xtab = pd.crosstab(\n",
- " df.end_period_base.replace(mapper),\n",
- " df.end_period_build.replace(mapper),\n",
+ " df.end_period_base.replace(mapper).fillna('Newly created'),\n",
+ " df.end_period_build.replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
")\n",
- "display(xtab.loc[tp_order,tp_order])"
+ "display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))"
]
},
{
@@ -1448,47 +1512,6 @@
"## At-Work Subtour Mode Choice"
]
},
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "0aa09d92-6245-4a8c-bbab-ec78cc62c4bf",
- "metadata": {},
- "outputs": [],
- "source": [
- "usecols = ['tour_id','tour_category','origin','destination','tour_category', 'tour_mode']\n",
- "base_tour = pd.read_csv(f\"{base_dir}/final_tours.csv\", \n",
- " index_col='tour_id', \n",
- " usecols=usecols)\n",
- "\n",
- "build_tour = pd.read_csv(f\"{build_dir}/final_tours.csv\", \n",
- " index_col='tour_id', \n",
- " usecols=usecols)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "c4fd7cb3-1920-4e62-b572-7325dee94ecd",
- "metadata": {},
- "outputs": [],
- "source": [
- "print(f\"\"\"Filtering tours by origin {\n",
- " 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
- " } destination MAZ.\\nOriginal tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\"\"\")\n",
- "base_tour = base_tour[multi_filter_mazs([base_tour.origin, base_tour.destination])] # base tour in the filtered set\n",
- "\n",
- "# special build-case filtering\n",
- "build_tour = build_tour[(\n",
- " build_tour.index.isin(base_tour_idx) # originally existed in base\n",
- " & build_tour.index.isin(base_tour.index) # and was in the current set for the base\n",
- " ) | # OR\n",
- " ((~build_tour.index.isin(base_tour_idx)) # is a new tour \n",
- " & multi_filter_mazs([build_tour.origin, build_tour.destination]) # and it's in this set\n",
- " )]\n",
- "\n",
- "print(f\"After filtering, tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\")"
- ]
- },
{
"cell_type": "code",
"execution_count": null,
@@ -1532,16 +1555,14 @@
"metadata": {},
"outputs": [],
"source": [
- "# xtab = df[[\"tour_mode_base\", \"tour_mode_build\"]].replace(combiners).value_counts(dropna=False).unstack().fillna(0)\n",
- "# xtab.loc[order,order]\n",
- "order = pd.Series(order)\n",
+ "\n",
"xtab = pd.crosstab(\n",
" df.tour_mode_base.replace(combiners).fillna('Newly created'),\n",
" df.tour_mode_build.replace(combiners).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total')\n",
"\n",
- "display(xtab.loc[order[order.isin(xtab.index)],order[order.isin(xtab.columns)]])"
+ "display(xtab.reindex(index=mode_order_base,columns=mode_order_build,fill_value=0))"
]
},
{
@@ -1567,184 +1588,6 @@
"\"\"\"\n",
"display(Markdown(summary_text.replace(\"\\n\",\"
\")))"
]
- },
- {
- "cell_type": "markdown",
- "id": "ac961eb2",
- "metadata": {},
- "source": [
- "## Stop Frequency"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "5eb7a6cd-6212-4503-9509-105802cff771",
- "metadata": {},
- "outputs": [],
- "source": [
- "usecols = ['tour_id','tour_category','origin','destination','stop_frequency', 'primary_purpose']\n",
- "base_tour = pd.read_csv(f\"{base_dir}/final_tours.csv\", \n",
- " index_col='tour_id', \n",
- " usecols=usecols)\n",
- "\n",
- "build_tour = pd.read_csv(f\"{build_dir}/final_tours.csv\", \n",
- " index_col='tour_id',\n",
- " usecols=usecols)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "f5698e6b-9d80-4398-90ae-21b55e3b049b",
- "metadata": {},
- "outputs": [],
- "source": [
- "print(f\"\"\"Filtering tours by origin {\n",
- " 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
- " } destination MAZ.\\nOriginal tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\"\"\")\n",
- "\n",
- "base_tour = base_tour[multi_filter_mazs([base_tour.origin, base_tour.destination])] # base tour in the filtered set\n",
- "\n",
- "# special build-case filtering\n",
- "build_tour = build_tour[(\n",
- " build_tour.index.isin(base_tour_idx) # originally existed in base\n",
- " & build_tour.index.isin(base_tour.index) # and was in the current set for the base\n",
- " ) | # OR\n",
- " ((~build_tour.index.isin(base_tour_idx)) # is a new tour \n",
- " & multi_filter_mazs([build_tour.origin, build_tour.destination]) # and it's in this set\n",
- " )]\n",
- "\n",
- "print(f\"After filtering, tours in base: {len(base_tour)}\\tbuild: {len(build_tour)}\")"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "317ed485-59ed-469d-9ba0-d6480e61c7e0",
- "metadata": {},
- "outputs": [],
- "source": [
- "df = pd.merge(\n",
- " base_tour,\n",
- " build_tour,\n",
- " how='outer',\n",
- " left_index=True,\n",
- " right_index=True,\n",
- " suffixes=['_base', '_build']\n",
- ")\n",
- "\n",
- "df.loc[df.stop_frequency_base == df.stop_frequency_build,'Case'] = \"Unchanged\"\n",
- "df.loc[df.stop_frequency_base != df.stop_frequency_build,'Case'] = \"Changed\"\n",
- "\n",
- "df.loc[df.stop_frequency_base.isna(),'Case'] = \"Newly Created\"\n",
- "df.loc[df.stop_frequency_build.isna(),\"Case\"] = \"Removed\"\n",
- "\n",
- "df.Case.value_counts()"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "38df880b",
- "metadata": {},
- "outputs": [],
- "source": [
- "total_stops_base = (df.stop_frequency_base.str[0].astype(float) \n",
- " + df.stop_frequency_base.str[5].astype(float)\n",
- " ).fillna(-np.inf).clip(upper=4)\n",
- "\n",
- "total_stops_build = (df.stop_frequency_build.str[0].astype(float) \n",
- " + df.stop_frequency_build.str[5].astype(float)\n",
- " ).fillna(-np.inf).clip(upper=4)\n",
- "\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "d9453567-16cd-47f1-b123-96ccaf0cd2ee",
- "metadata": {},
- "outputs": [],
- "source": [
- "fig = vh.create_pie_chart(df.Case.value_counts().to_frame().sort_index(), [\"count\"])\n",
- "fig"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "db5a41d1-123e-43b1-b5c9-7f32009c2dbe",
- "metadata": {},
- "outputs": [],
- "source": [
- "diff = total_stops_build - total_stops_base\n",
- "xtab = pd.crosstab(\n",
- " total_stops_base.replace({4:\"4+\",-np.inf:\"Newly created\"}).astype(str).str.replace(\".0\",\"\"),\n",
- " total_stops_build.clip(upper=4).replace({4:\"4+\",-np.inf:\"Removed\"}).astype(str).str.replace(\".0\",\"\"),\n",
- " margins=True,\n",
- " margins_name='Total')\n",
- "\n",
- "display(xtab.sort_index())"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "7f6eb5e6",
- "metadata": {},
- "outputs": [],
- "source": [
- "summary_text = f\"\"\"**Stop frequency changes:**\n",
- " Tours with fewer stops in build: {len(diff[diff < 0])}\n",
- " Tours with the same stops in build: {len(diff[diff==0])}\n",
- " Tours with more stops in build: {len(diff[diff > 0])}\n",
- "\"\"\"\n",
- "display(Markdown(summary_text.replace(\"\\n\",\"
\")))"
- ]
- },
- {
- "cell_type": "raw",
- "id": "4115c2e0-dc14-410f-8716-457c78db6e70",
- "metadata": {},
- "source": [
- "::: {.panel-tabset}"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "536eca88-0f30-4fa3-a463-802fa3da2fb0",
- "metadata": {},
- "outputs": [],
- "source": [
- "# | output: asis\n",
- "# above comment is needed for Quarto to render subtabs correctly\n",
- "\n",
- "# only trips that exist in the base run will be output\n",
- "# looping thru name_dict will not find cases where the base value for purpose or category is nan in the outer join\n",
- "for key, value in name_dict.items():\n",
- " df_purp = df.loc[df.primary_purpose_base == key]\n",
- " if len(df_purp) == 0:\n",
- " continue \n",
- " \n",
- " print(f\"#### {value}\")\n",
- " df_purp_cases = df_purp.Case.value_counts()\n",
- " print(str(df_purp_cases).replace(\"\\n\",\"
\"))\n",
- " \n",
- " fig = vh.create_pie_chart(df_purp_cases.to_frame().sort_index(),[\"count\"])\n",
- " \n",
- " fig.show()\n",
- " display(Markdown(\" \"))"
- ]
- },
- {
- "cell_type": "raw",
- "id": "1da243b8-b557-427f-82a0-3928c653f0bb",
- "metadata": {},
- "source": [
- ":::"
- ]
}
],
"metadata": {
diff --git a/notebooks/trips.ipynb b/notebooks/trips.ipynb
index 4e5a9ed..2eef78e 100644
--- a/notebooks/trips.ipynb
+++ b/notebooks/trips.ipynb
@@ -103,7 +103,7 @@
"metadata": {},
"outputs": [],
"source": [
- "print(f\"\"\"Filtering trips by origin {\n",
+ "print(f\"\"\"Filtering all trips by origin {\n",
" 'and' if how_method == 'all' else 'or' if how_method == 'any' else '???'\n",
" } destination MAZ.\\nOriginal trips in base: {len(base_trip)}\\tbuild: {len(build_trip)}\\tdiff: {len(build_trip)-len(base_trip)}\"\"\")\n",
"\n",
@@ -120,7 +120,7 @@
" \n",
" ]\n",
"\n",
- "print(f\"After filtering, trips in base: {len(base_trip)}\\tbuild: {len(build_trip)}\\tdiff: {len(build_trip)-len(base_trip)}\")"
+ "print(f\"After filtering, total trips in base: {len(base_trip)}\\tbuild: {len(build_trip)}\\tdiff: {len(build_trip)-len(base_trip)}\")"
]
},
{
@@ -128,7 +128,7 @@
"id": "265516ee",
"metadata": {},
"source": [
- "## Trip Purpose"
+ "## Intermediate Stop Purpose"
]
},
{
@@ -192,7 +192,7 @@
"id": "6a7d1649",
"metadata": {},
"source": [
- "## Trip Destination"
+ "## Intermediate Stop Destination"
]
},
{
@@ -269,7 +269,7 @@
"# compare trips that are in both base and build\n",
"df = base_trip.merge(\n",
" build_trip, \n",
- " how='inner', \n",
+ " how='outer', \n",
" left_index=True, \n",
" right_index=True,\n",
" suffixes=('_base', '_build')\n",
@@ -295,18 +295,19 @@
" 3: \"PM\",\n",
" 4: \"EV\"\n",
"}\n",
- "tp_order = ['EA','AM','MD','PM','EV','Total']\n",
+ "tp_order_base = ['EA','AM','MD','PM','EV','Newly created','Total']\n",
+ "tp_order_build = ['EA','AM','MD','PM','EV','Removed','Total']\n",
"\n",
"metric='depart'\n",
"purpose_df = df\n",
- "display(Markdown(f\"### Trip departure changes\"))\n",
+ "display(Markdown(f\"### Trip departure changes for all trips\"))\n",
"xtab = pd.crosstab(\n",
- " purpose_df[f'{metric}_period_base'].replace(mapper),\n",
- " purpose_df[f'{metric}_period_build'].replace(mapper),\n",
+ " purpose_df[f'{metric}_period_base'].replace(mapper).fillna('Newly created'),\n",
+ " purpose_df[f'{metric}_period_build'].replace(mapper).fillna('Removed'),\n",
" margins=True,\n",
" margins_name='Total'\n",
")\n",
- "display(xtab.loc[tp_order,tp_order])\n",
+ "display(xtab.reindex(index=tp_order_base,columns=tp_order_build,fill_value=0))\n",
"\n",
"summary_text = f\"\"\"**Trip departure changes:**\n",
" Trips which departed earlier in build: {len(purpose_df[purpose_df[f'{metric}_bin_difference'] < 0])}\n",
@@ -429,7 +430,7 @@
"metadata": {},
"outputs": [],
"source": [
- "summary_text = f\"\"\"**Trip mode changes:**\n",
+ "summary_text = f\"\"\"**Trip mode changes for all trips:**\n",
"To transit\n",
"\n",
"from auto: {len(df[df.trip_mode_base.isin(auto_modes) & df.trip_mode_build.isin(transit_modes)])}\n",
diff --git a/visualizer_helpers.py b/visualizer_helpers.py
index a968f69..d246a87 100644
--- a/visualizer_helpers.py
+++ b/visualizer_helpers.py
@@ -34,6 +34,8 @@
"Removed": "Cherry_light",
"Increased": "Leaf",
"Decreased": "Cherry",
+ "Workplace moved into area": "Leaf_light",
+ "Workplace moved out of area": "Cherry_light",
}
def create_bar_chart(
@@ -252,7 +254,7 @@ def unaffected_multi(xs: pd.Series) -> pd.Series:
def get_time_period_index(
bin: int,
- time_period_mapping: Optional[list] = [0,12,22,32,40,48],
+ time_period_mapping: Optional[list] = [0,6,12,25,32,48],
) -> int:
"""
Convert a bin number to a time period index.