-
Notifications
You must be signed in to change notification settings - Fork 14
/
agg.py
489 lines (390 loc) · 21.6 KB
/
agg.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
import numpy as np
import pandas as pd
import h5py
import os, sys, shutil
CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
sys.path.append(os.path.dirname(CURRENT_DIR))
sys.path.append(os.path.join(os.getcwd(),"inputs"))
sys.path.append(os.path.join(os.getcwd(),"scripts"))
sys.path.append(os.getcwd())
import re
import math
from collections import OrderedDict
# from input_configuration import base_year
import time
import toml
config = toml.load(os.path.join(os.getcwd(), 'configuration/input_configuration.toml'))
# Define relationships between daysim files
daysim_merge_fields = {'Trip':
{'Tour': ['hhno','pno','tour'],
'Person': ['hhno','pno'],
'Household': ['hhno']},
'Person':
{'Household': ['hhno']},
'Tour':
{'Person': ['hhno','pno'],
'Household': ['hhno']
}
}
dash_table_list = ['vmt_facility','vht_facility','delay_facility','vmt_user_class','vht_user_class','delay_user_class']
def get_dict_values(d):
"""Return unique dictionary values for a 2-level dictionary"""
_list = []
for k, v in d.iteritems():
if isinstance(v, dict):
for _k, _v in v.iteritems():
_list += _v
else:
_list += v
_list = list(np.unique(_list))
return(_list)
def create_dir(_dir):
if os.path.exists(_dir):
shutil.rmtree(_dir)
os.makedirs(_dir)
def get_row_col_list(row, full_col_list):
row_list = ['agg_fields','values']
for field_type in ['filter_fields']:
if type(row[field_type]) != np.float:
row_list += [field_type]
col_list = list(row[row_list].values)
col_list = [i.split(',') for i in col_list]
col_list = list(np.unique([item.strip(' ') for sublist in col_list for item in sublist]))
# Identify column values from query field with regular expressions
if type(row['query']) != np.float:
# query_fields_cols = [i.strip() for i in re.split(',|>|==|>=|<|<=|!=|&',row['query'])]
regex = re.compile('[^a-zA-Z]')
query_fields_cols = [regex.sub('', i).strip() for i in re.split(',|>|==|>=|<|<=|!=|&',row['query'])]
for query in query_fields_cols:
if query in full_col_list and query not in col_list:
col_list += [query]
return col_list
def merge_geography(df, df_geog, parcel_geog, buffered_parcels):
for _index, _row in df_geog.iterrows():
right_df = pd.eval(_row['right_table'] + "[" + str(list(_row[['right_index','right_column']].values)) + "]", engine='python')
df = df.merge(right_df, left_on=_row['left_index'], right_on=_row['right_index'], how='left')
df.rename(columns={_row['right_column']: _row['right_column_rename']}, inplace=True)
df.drop(_row['right_index'], axis=1, inplace=True)
return df
def execute_eval(df, row, col_list, fname):
# Process query field
query = ''
if type(row['query']) != np.float:
query = """.query('"""+ str(row['query']) + """')"""
agg_fields_cols = [i.strip() for i in row['agg_fields'].split(',')]
values_cols = [i.strip() for i in row['values'].split(',')]
if type(row['filter_fields']) == np.float:
expr = 'df[' + str(col_list) + ']' + query + ".groupby(" + str(agg_fields_cols) + ")." + row['aggfunc'] + "()[" + str(values_cols) + "]"
# Write results to target output
df_out = pd.eval(expr, engine='python').reset_index()
_labels_df = labels_df[labels_df['field'].isin(df_out.columns)]
for field in _labels_df['field'].unique():
_df = _labels_df[_labels_df['field'] == field]
local_series = pd.Series(_df['text'].values, index=_df['value'])
df_out[field] = df_out[field].map(local_series)
df_out.to_csv(fname+'.csv', index=False)
else:
filter_cols = np.unique([i.strip() for i in row['filter_fields'].split(',')])
for _filter in filter_cols:
unique_vals = np.unique(df[_filter].values.astype('str'))
for filter_val in unique_vals:
expr = 'df[' + str(col_list) + "][df['" + str(_filter) + "'] == '" + str(filter_val) + "']" + \
".groupby(" + str(agg_fields_cols) + ")." + row['aggfunc'] + "()[" + str(values_cols) + "]"
# Write results to target output
df_out = pd.eval(expr, engine='python').reset_index()
# # Apply labels
_labels_df = labels_df[labels_df['field'].isin(df_out.columns)]
for field in _labels_df['field'].unique():
_df = _labels_df[_labels_df['field'] == field]
local_series = pd.Series(_df['text'].values, index=_df['value'])
df_out[field] = df_out[field].map(local_series)
df_out.to_csv(fname+'_'+str(_filter)+'_'+str(filter_val)+'.csv', index=False)
def h5_df(h5file, table, col_list):
df = pd.DataFrame()
for col in col_list:
df[col] = h5file[table][col][:].astype('float32')
return df
def create_agg_outputs(path_dir_base, base_output_dir, survey=False):
# Load the expression file
expr_df = pd.read_csv(os.path.join(os.getcwd(),r'inputs/model/summaries/agg_expressions.csv'))
# expr_df = expr_df.fillna('__remove__') # Fill NA with string signifying data to be ignored
geography_lookup = pd.read_csv(os.path.join(os.getcwd(),r'inputs/model/summaries/geography_lookup.csv'))
variables_df = pd.read_csv(os.path.join(os.getcwd(),r'inputs/model/summaries/variables.csv'))
global labels_df
labels_df = pd.read_csv(os.path.join(os.getcwd(),'inputs/model/lookup/variable_labels.csv'))
geog_cols = list(np.unique(geography_lookup[geography_lookup.right_table == 'parcel_geog'][['right_column','right_index']].values))
# Add geographic lookups at parcel level; only load relevant columns
parcel_geog = pd.read_sql_table('parcel_'+config['base_year']+'_geography', 'sqlite:///inputs/db/soundcast_inputs.db',
columns=geog_cols)
buffered_parcels_cols = list(np.unique(geography_lookup[geography_lookup.right_table == 'buffered_parcels'][['right_column','right_index']]))
buffered_parcels = pd.read_csv(os.path.join(os.getcwd(),r'outputs/landuse/buffered_parcels.txt'), delim_whitespace=True,
usecols=buffered_parcels_cols)
# Create output folder for flattened output
if survey:
survey_str = 'survey'
# Get a list of headers for all daysim records so we can load data in as needed
else:
survey_str = ''
# create h5 table of daysim outputs
daysim_h5 = h5py.File(os.path.join(path_dir_base, 'daysim_outputs.h5'), 'r')
# daysim_h5 = h5py.File()
for dirname in pd.unique(expr_df['output_dir']):
if survey:
create_dir(os.path.join(base_output_dir,dirname,'survey'))
else:
create_dir(os.path.join(base_output_dir,dirname))
# Expression log
expr_log_path = r'outputs/agg/expr_log.csv'
if os.path.exists(expr_log_path):
os.remove(expr_log_path)
hh_full_col_list = pd.read_csv(os.path.join(path_dir_base,'_household.tsv'), delim_whitespace=True, nrows=0)
person_full_col_list = pd.read_csv(os.path.join(path_dir_base,'_person.tsv'), delim_whitespace=True, nrows=0)
trip_full_col_list = pd.read_csv(os.path.join(path_dir_base,'_trip.tsv'), delim_whitespace=True, nrows=0)
tour_full_col_list = pd.read_csv(os.path.join(path_dir_base,'_tour.tsv'), delim_whitespace=True, nrows=0)
var_list = list(variables_df['new_variable'])
# Full list of potential columns
full_col_list = list(hh_full_col_list) + list(person_full_col_list) + list(trip_full_col_list) + list(tour_full_col_list) + geog_cols + var_list
#####################
## Household
#####################
df_agg = expr_df[expr_df['table'] == 'household']
# Loop through each expression and evaluat result
# only load the necessary columns and data for this row
for index, row in df_agg.iterrows():
data_tables = {}
col_list = get_row_col_list(row, full_col_list)
# load the required data for the main df (houeshold)
load_cols = [i for i in col_list if i in hh_full_col_list] + ['hhparcel']
# Also account for any added user variables
user_var_cols = [i for i in col_list if i in variables_df['new_variable'].values]
if len(user_var_cols) > 0:
df_var = variables_df[variables_df['new_variable'].isin(col_list)]
load_cols += df_var['modified_variable'].values.tolist()
# Don't load any variables from buffered_parcels_cols
_buffered_cols = []
for i in load_cols:
if i in buffered_parcels_cols:
load_cols.remove(i)
_buffered_cols.append(i)
if survey:
household = pd.read_csv(os.path.join(path_dir_base,'_household.tsv'), delim_whitespace=True, usecols=load_cols)
else:
household = h5_df(daysim_h5, 'Household', load_cols)
# merge geography and other variables
geog_cols = [i for i in col_list if i in geography_lookup['right_column_rename'].values]
if len(geog_cols) > 0:
df_geog = geography_lookup[geography_lookup['right_column_rename'].isin(col_list)]
household = merge_geography(household, df_geog, parcel_geog, buffered_parcels)
# Account for any buffered parcel cols used in the data
if not set(col_list).isdisjoint(buffered_parcels_cols) or len(_buffered_cols) > 0:
household = merge_geography(household, geography_lookup[(geography_lookup.right_table == 'buffered_parcels') &
(geography_lookup.left_table == 'Household')], parcel_geog, buffered_parcels)
# Calculate user variables
if len(user_var_cols) > 0:
# df_var = variables_df[variables_df['new_variable'].isin(col_list)]
for _index, _row in df_var.iterrows():
household[_row['new_variable']] = eval(_row['expression'])
del df_var
fname = os.path.join(base_output_dir, str(row['output_dir']),survey_str,str(row['target']))
execute_eval(household, row, col_list, fname)
del [household]
#################################
# Person
#################################
df_agg = expr_df[expr_df['table'] == 'person']
# Loop through each expression and evaluat result
# only load the necessary columns and data for this row
for index, row in df_agg.iterrows():
data_tables = {}
col_list = get_row_col_list(row, full_col_list)
# load the required data for the main df (houeshold)
load_cols = [i for i in col_list if i in person_full_col_list] + ['hhno','pwpcl','psexpfac']
# Also account for any added user variables
user_var_cols = [i for i in col_list if i in variables_df['new_variable'].values]
if len(user_var_cols) > 0:
df_var = variables_df[variables_df['new_variable'].isin(col_list)]
load_cols += df_var['modified_variable'].values.tolist()
# Don't load any variables from buffered_parcels_cols
_buffered_cols = []
for i in load_cols:
if i in buffered_parcels_cols:
load_cols.remove(i)
_buffered_cols.append(i)
# also load any columns needed for geographic join
df_geog = geography_lookup[geography_lookup['left_table'] == 'Person']
df_geog = df_geog[df_geog['right_column_rename'].isin(col_list)]
load_cols += df_geog['left_index'].values.tolist()
if survey:
person = pd.read_csv(os.path.join(path_dir_base,'_person.tsv'), delim_whitespace=True, usecols=load_cols)
else:
person = h5_df(daysim_h5, 'Person', load_cols)
# households
# Also account for any added user variables
load_cols = [i for i in col_list if i in hh_full_col_list] + ['hhno','hhparcel']
if survey:
household = pd.read_csv(os.path.join(path_dir_base,'_household.tsv'), delim_whitespace=True, usecols=load_cols)
else:
household = h5_df(daysim_h5, 'Household', load_cols)
if len(household) > 0:
person = person.merge(household, on='hhno')
# merge geography and other variables
geog_cols = [i for i in col_list if i in geography_lookup['right_column_rename'].values]
if len(geog_cols) > 0:
df_geog = geography_lookup[geography_lookup['right_column_rename'].isin(col_list)]
person = merge_geography(person, df_geog, parcel_geog, buffered_parcels)
# Account for any buffered parcel cols used in the data
if not set(col_list).isdisjoint(buffered_parcels_cols) or len(_buffered_cols) > 0:
person = merge_geography(person, geography_lookup[(geography_lookup.right_table == 'buffered_parcels') &
(geography_lookup.left_table == 'Person')], parcel_geog, buffered_parcels)
# Calculate user variables
if len(user_var_cols) > 0:
for _index, _row in df_var.iterrows():
person[_row['new_variable']] = pd.eval(_row['expression'],engine='python')
del df_var
fname = os.path.join(base_output_dir, str(row['output_dir']),survey_str,str(row['target']))
execute_eval(person, row, col_list, fname)
del [household, person]
#################################
# Trips
#################################
df_agg = expr_df[expr_df['table'] == 'trip']
# Loop through each expression and evaluate result
# only load the necessary columns and data for this row
for index, row in df_agg.iterrows():
data_tables = {}
col_list = get_row_col_list(row, full_col_list)
# households
# Also account for any added user variables
load_cols = [i for i in col_list if i in hh_full_col_list] + ['hhno','hhparcel']
if survey:
household = pd.read_csv(os.path.join(path_dir_base,'_household.tsv'), delim_whitespace=True, usecols=load_cols)
else:
household = h5_df(daysim_h5, 'Household', load_cols)
# persons
# Also account for any added user variables
load_cols = [i for i in col_list if i in person_full_col_list] + ['hhno','pno','psexpfac']
if survey:
person = pd.read_csv(os.path.join(path_dir_base,'_person.tsv'), delim_whitespace=True, usecols=load_cols)
else:
person = h5_df(daysim_h5, 'Person', load_cols)
# trips
# Also account for any added user variables
load_cols = list(np.unique([i for i in col_list if i in trip_full_col_list] + ['pno','hhno','tour','trexpfac']))
# only load user variables that are related to this table
user_var_cols = [i for i in col_list if i in variables_df['new_variable'].values]
if len(user_var_cols) > 0:
df_var = variables_df[variables_df['new_variable'].isin(col_list)]
load_cols += df_var['modified_variable'].values.tolist()
# also load any columns needed for geographic join
geog_cols = [i for i in col_list if i in geography_lookup['right_column_rename'].values]
if len(geog_cols) > 0:
df_geog = geography_lookup[geography_lookup['left_table'] == 'Trip']
df_geog = df_geog[df_geog['right_column_rename'].isin(col_list)]
load_cols += df_geog['left_index'].values.tolist()
if survey:
trip = pd.read_csv(os.path.join(path_dir_base,'_trip.tsv'), delim_whitespace=True, usecols=load_cols)
else:
trip = h5_df(daysim_h5, 'Trip', load_cols)
# tours
# Also account for any added user variables
load_cols = [i for i in col_list if i in tour_full_col_list] + ['pno','hhno','tour']
tour = pd.read_csv(os.path.join(path_dir_base,'_tour.tsv'), delim_whitespace=True, usecols=load_cols)
# merge geography and other variables
geog_cols = [i for i in col_list if i in geography_lookup['right_column_rename'].values]
if len(geog_cols) > 0:
trip = merge_geography(trip, df_geog, parcel_geog, buffered_parcels)
# merge geography based on household info
df_geog = geography_lookup[geography_lookup['left_table'] == 'Household']
df_geog = df_geog[df_geog['right_column_rename'].isin(col_list)]
if len(geog_cols) > 0:
household = merge_geography(household, df_geog, parcel_geog, buffered_parcels)
trip = trip.merge(household, on=['hhno'])
if len([i for i in col_list if i in person_full_col_list]) > 0:
trip = trip.merge(person, on=['hhno','pno'])
if len([i for i in col_list if i in tour_full_col_list]) > 0:
trip = trip.merge(tour, on=['pno','hhno','tour'])
# Calculate user variables
if len(user_var_cols) > 0:
for _index, _row in df_var.iterrows():
trip[_row['new_variable']] = pd.eval(_row['expression'],engine='python')
fname = os.path.join(base_output_dir, str(row['output_dir']),survey_str,str(row['target']))
execute_eval(trip, row, col_list, fname)
del [tour, trip, person, household, df_geog]
############################
# Tour
############################
df_agg = expr_df[expr_df['table'] == 'tour']
# Loop through each expression and evaluate result
# only load the necessary columns and data for this row
for index, row in df_agg.iterrows():
data_tables = {}
col_list = get_row_col_list(row, full_col_list)
# households
# Also account for any added user variables
load_cols = [i for i in col_list if i in hh_full_col_list] + ['hhno','hhparcel']
if survey:
household = pd.read_csv(os.path.join(path_dir_base,'_household.tsv'), delim_whitespace=True, usecols=load_cols)
else:
household = h5_df(daysim_h5, 'Household', load_cols)
# persons
# Also account for any added user variables
load_cols = [i for i in col_list if i in person_full_col_list] + ['hhno','pno']
if survey:
person = pd.read_csv(os.path.join(path_dir_base,'_person.tsv'), delim_whitespace=True, usecols=load_cols)
else:
person = h5_df(daysim_h5, 'Person', load_cols)
# tours
# Also account for any added user variables
load_cols = list(np.unique([i for i in col_list if i in tour_full_col_list] + ['pno','hhno','tour','toexpfac']))
# only load user variables that are related to this table
user_var_cols = [i for i in col_list if i in variables_df['new_variable'].values]
if len(user_var_cols) > 0:
df_var = variables_df[variables_df['new_variable'].isin(col_list)]
load_cols += df_var['modified_variable'].values.tolist()
# also load any columns needed for geographic join
geog_cols = [i for i in col_list if i in geography_lookup['right_column_rename'].values]
if len(geog_cols) > 0:
df_geog = geography_lookup[geography_lookup['left_table'] == 'Tour']
df_geog = df_geog[df_geog['right_column_rename'].isin(col_list)]
load_cols += df_geog['left_index'].values.tolist()
if survey:
tour = pd.read_csv(os.path.join(path_dir_base,'_tour.tsv'), delim_whitespace=True, usecols=load_cols)
else:
tour = h5_df(daysim_h5, 'Tour', load_cols)
# merge geography and other variables
geog_cols = [i for i in col_list if i in geography_lookup['right_column_rename'].values]
df_geog = geography_lookup[geography_lookup['left_table'] == 'Tour']
df_geog = df_geog[df_geog['right_column_rename'].isin(col_list)]
if len(geog_cols) > 0:
tour = merge_geography(tour, df_geog, parcel_geog, buffered_parcels)
# merge geography based on household info
df_geog = geography_lookup[geography_lookup['left_table'] == 'Household']
df_geog = df_geog[df_geog['right_column_rename'].isin(col_list)]
if len(geog_cols) > 0:
household = merge_geography(household, df_geog, parcel_geog, buffered_parcels)
tour = tour.merge(household, on=['hhno'])
if len(person) > 0:
tour = tour.merge(person, on=['hhno','pno'])
# Calculate user variables
if len(user_var_cols) > 0:
for _index, _row in df_var.iterrows():
tour[_row['new_variable']] = pd.eval(_row['expression'],engine='python')
fname = os.path.join(base_output_dir, str(row['output_dir']),survey_str,str(row['target']))
execute_eval(tour, row, col_list, fname)
del [tour, person, household, df_geog]
def copy_dash_tables(dash_table_list):
"""Copy outputs from validation and network_summary scripts required for Dash."""
for fname in dash_table_list:
shutil.copy(os.path.join(r'outputs/network',fname+'.csv'), r'outputs/agg/dash')
def main():
output_dir_base = os.path.join(os.getcwd(),'outputs/agg')
create_dir(output_dir_base)
input_dir = os.path.join(os.getcwd(),r'outputs/daysim')
create_agg_outputs(input_dir, output_dir_base, survey=False)
survey_input_dir = os.path.join(os.getcwd(),r'inputs/base_year/survey')
create_agg_outputs(survey_input_dir, output_dir_base, survey=True)
copy_dash_tables(dash_table_list)
if __name__ == '__main__':
start_time = time.time()
main()
print("--- %s seconds ---" % (time.time() - start_time))