Make handling of EMME matrices less error-prone and more flexible #525

zptro · 2023-06-20T08:47:13Z

With this PR, EMME matrix ids are no longer set manually. Instead, they are created with an algorithm according to the same principles as before:

The id number space is one hundred for each time period (if matrices for each time period is to be saved in EMME).
The id number space is ten for each matrix type (demand, time, dist, cost, gen_cost), but the algorithm checks whether the number of assignment classes is more than ten; in that case it assigns more than ten numbers per matrix type.
As previously, the extra transit impedance matrices (that are not used in demand calculation or cba), are stored in a separate number space with 10 numbers for each transit assignment class. The calculation of these is no longer hard-coded in TransitSpecification; instead the appropriate matrix names are taken from the parameter file.

The matrix id numbers are now internally stored in a more simple dict structure (which directly corresponds to how the matrices are handled in the scripts), which is created in a separate method:

helmet-model-system/Scripts/assignment/emme_assignment.py

Lines 337 to 344 in d8fc46b

    
                   dict 
        
                       key : str 
        
                           Assignment class (car_work/transit_leisure/...) 
        
                       value : dict 
        
                           key : str 
        
                               Matrix type (demand/time/cost/dist/...) 
        
                           value : str 
        
                               EMME matrix id

Also add docstrings and remove deprecated imports

zptro · 2023-06-29T06:16:53Z

@hsl-petrhaj @johpiip are you able to look at this?

hsl-petrhaj · 2023-06-29T06:55:58Z

I did not have time for testing it yet, I should be able to check it this week.

hsl-petrhaj

Testing looks fine with mock run, so I approve it for now.

hsl-petrhaj · 2023-06-29T07:23:47Z

Now testing still with Emme run.

hsl-petrhaj · 2023-06-29T07:51:30Z

@zptro The validate script (helmet_validate_inputfiles.py) now does not work, as the parameter names have changed. Please also check the tests.

[UI-event] Initializing run of scenarios: kth_testi2
[INFO 10:47] Checking base inputdata...
[INFO 10:47] Checking base zonedata & scenario-based input data...
[INFO 10:47] Checking input data for scenario #0 ...
[EXCEPTION] Traceback (most recent call last):
[EXCEPTION]   File "...\Scripts\helmet_validate_inputfiles.py", line 252, in <module>
[EXCEPTION]     main(args)
[EXCEPTION]   File "...\Scripts\helmet_validate_inputfiles.py", line 135, in main
[EXCEPTION]     nr_vehicle_classes = len(param.emme_demand_mtx) + 1
[EXCEPTION] AttributeError: module 'parameters.assignment' has no attribute 'emme_demand_mtx'
[ERROR] {"message": "Checking base inputdata...", "level": "INFO"}
{"message": "Checking base zonedata & scenario-based input data...", "level": "INFO"}
{"message": "Checking input data for scenario #0 ...", "level": "INFO"}
Traceback (most recent call last):
  File "...\Scripts\helmet_validate_inputfiles.py", line 252, in <module>
    main(args)
  File "...\Scripts\helmet_validate_inputfiles.py", line 135, in main
    nr_vehicle_classes = len(param.emme_demand_mtx) + 1
AttributeError: module 'parameters.assignment' has no attribute 'emme_demand_mtx'

hsl-petrhaj · 2023-06-29T11:24:53Z

For some reason I had to increase the size of the Emmebank for extra attributes, I guess we have some new ones now? I will then need to update helmet-docs as well.

hsl-petrhaj

Should be fine now. Not sure why I got the "not enough space" error in Emme, but perhaps this could be coming from somewhere else. Anyway it was easy to get this version to run.

zptro · 2023-06-29T12:03:53Z

For some reason I had to increase the size of the Emmebank for extra attributes, I guess we have some new ones now? I will then need to update helmet-docs as well.

That is not very good. I found the problem and fixed. We do not need an extra attribute for walking, as we do not assign walk trips.

hsl-petrhaj · 2023-06-29T13:37:23Z

I am getting some division by zero errors when running with Emme in the second iteration. Trying to figure out if it is related to those changes.

hsl-petrhaj · 2023-06-29T13:53:03Z

@zptro
I just verified that the error only appears in this pull request. Here is the error:


[INFO 16:50] Starting iteration 1
[INFO 16:50] Demand calculation started...
[EXCEPTION] ...\Scripts\models\logit.py:401: RuntimeWarning: divide by zero encountered in log
[EXCEPTION]   logsum = pandas.Series(numpy.log(expsum), self.purpose.zone_numbers)
[EXCEPTION] ...\Scripts\models\logit.py:418: RuntimeWarning: invalid value encountered in true_divide
[EXCEPTION]   / self.dest_expsums[mode]["logsum"])
[EXCEPTION] ...\Scripts\datatypes\purpose.py:258: RuntimeWarning: invalid value encountered in greater
[EXCEPTION]   dests = generation > secondary_destination_threshold
[INFO 16:50] Demand calculation completed
[WARN] (2, 12) car_leisure matrix not matching 28 demand shares. Resorted to backup demand shares.
[WARN] (2, 12) car_leisure matrix not matching 28 demand shares. Resorted to backup demand shares.
[EXCEPTION] ...\modelsystem.py:289: RuntimeWarning: invalid value encountered in double_scalars
[EXCEPTION]   mode_shares[mode] = tour_sum[mode].sum() / sum_all.sum()
[INFO 16:50] Assigning period aht
[ERROR] NAs in demand matrix car_leisure would cause infinite loop in Emme assignment.
[ERROR] Exception at iteration 1
[EXCEPTION] Traceback (most recent call last):
[EXCEPTION]   File "...\Scripts\helmet.py", line 105, in main
[EXCEPTION]     else model.run_iteration(impedance, i))
[EXCEPTION]   File "...\Scripts\modelsystem.py", line 310, in run_iteration
[EXCEPTION]     impedance[tp] = ap.assign(self.dtm.demand[tp], iteration)
[EXCEPTION]   File "...\Scripts\assignment\assignment_period.py", line 107, in assign
[EXCEPTION]     self._set_emmebank_matrices(matrices, iteration=="last")
[EXCEPTION]   File "...\Scripts\assignment\assignment_period.py", line 417, in _set_emmebank_matrices
[EXCEPTION]     self._set_matrix(mtx, matrices[mtx])
[EXCEPTION]   File "...\Scripts\assignment\assignment_period.py", line 424, in _set_matrix
[EXCEPTION]     raise ValueError(msg)
[EXCEPTION] ValueError: NAs in demand matrix car_leisure would cause infinite loop in Emme assignment.
[EXCEPTION] {"message": "Fatal error occured, simulation aborted.", "level": "ERROR", "status": {"name": "test100", "state": "running", "current": 1, "completed": 0, "failed": 1, "total": 15, "log": "E:\\Results\\test0\\test100\\test100.log", "converged": 0, "results": [{"car": NaN, "transit": NaN, "bike": NaN, "walk": NaN}]}}
[WARN] Model has not converged
[INFO 16:50] Removed strategy files in D:\helmet_41\Projektit\testiverkko\database
[EXCEPTION] {"message": "Simulation ended.", "level": "INFO", "status": {"name": "test100", "state": "running", "current": 1, "completed": 0, "failed": 1, "total": 15, "log": "E:\\Results\\test0\\test100\\test100.log", "converged": 0, "results": [{"car": NaN, "transit": NaN, "bike": NaN, "walk": NaN}]}}
[UI-event] Scenario "test100" finished

Could there be some issue with the demand matrix?

hsl-petrhaj

Still needs some further checks.

zptro · 2023-08-01T13:30:07Z

@zptro I just verified that the error only appears in this pull request. Here is the error:


[INFO 16:50] Starting iteration 1
[INFO 16:50] Demand calculation started...
[EXCEPTION] ...\Scripts\models\logit.py:401: RuntimeWarning: divide by zero encountered in log
[EXCEPTION]   logsum = pandas.Series(numpy.log(expsum), self.purpose.zone_numbers)
[EXCEPTION] ...\Scripts\models\logit.py:418: RuntimeWarning: invalid value encountered in true_divide
[EXCEPTION]   / self.dest_expsums[mode]["logsum"])
[EXCEPTION] ...\Scripts\datatypes\purpose.py:258: RuntimeWarning: invalid value encountered in greater
[EXCEPTION]   dests = generation > secondary_destination_threshold
[INFO 16:50] Demand calculation completed
[WARN] (2, 12) car_leisure matrix not matching 28 demand shares. Resorted to backup demand shares.
[WARN] (2, 12) car_leisure matrix not matching 28 demand shares. Resorted to backup demand shares.
[EXCEPTION] ...\modelsystem.py:289: RuntimeWarning: invalid value encountered in double_scalars
[EXCEPTION]   mode_shares[mode] = tour_sum[mode].sum() / sum_all.sum()
[INFO 16:50] Assigning period aht
[ERROR] NAs in demand matrix car_leisure would cause infinite loop in Emme assignment.
[ERROR] Exception at iteration 1
[EXCEPTION] Traceback (most recent call last):
[EXCEPTION]   File "...\Scripts\helmet.py", line 105, in main
[EXCEPTION]     else model.run_iteration(impedance, i))
[EXCEPTION]   File "...\Scripts\modelsystem.py", line 310, in run_iteration
[EXCEPTION]     impedance[tp] = ap.assign(self.dtm.demand[tp], iteration)
[EXCEPTION]   File "...\Scripts\assignment\assignment_period.py", line 107, in assign
[EXCEPTION]     self._set_emmebank_matrices(matrices, iteration=="last")
[EXCEPTION]   File "...\Scripts\assignment\assignment_period.py", line 417, in _set_emmebank_matrices
[EXCEPTION]     self._set_matrix(mtx, matrices[mtx])
[EXCEPTION]   File "...\Scripts\assignment\assignment_period.py", line 424, in _set_matrix
[EXCEPTION]     raise ValueError(msg)
[EXCEPTION] ValueError: NAs in demand matrix car_leisure would cause infinite loop in Emme assignment.
[EXCEPTION] {"message": "Fatal error occured, simulation aborted.", "level": "ERROR", "status": {"name": "test100", "state": "running", "current": 1, "completed": 0, "failed": 1, "total": 15, "log": "E:\\Results\\test0\\test100\\test100.log", "converged": 0, "results": [{"car": NaN, "transit": NaN, "bike": NaN, "walk": NaN}]}}
[WARN] Model has not converged
[INFO 16:50] Removed strategy files in D:\helmet_41\Projektit\testiverkko\database
[EXCEPTION] {"message": "Simulation ended.", "level": "INFO", "status": {"name": "test100", "state": "running", "current": 1, "completed": 0, "failed": 1, "total": 15, "log": "E:\\Results\\test0\\test100\\test100.log", "converged": 0, "results": [{"car": NaN, "transit": NaN, "bike": NaN, "walk": NaN}]}}
[UI-event] Scenario "test100" finished

Could there be some issue with the demand matrix?

The problem was that transit time matrices were 9999, because assignment did not write anything to them, because I had accidentally dropped that line. Now this is fixed.

zptro · 2023-08-01T13:31:10Z

@johpiip could you review this now?

johpiip

Hi and thank you for your PR! I am sad to see the Emme matrix cheatsheet go but I understand the reasoning behind. 😊

I tried running this branch with the samll test network. I encountered an type error, maybe you can check it?

[UI-event] Initializing run of scenarios: testi_emme-matrices
[INFO 10:21] Checking base inputdata...
[INFO 10:21] Checking base zonedata & scenario-based input data...
[INFO 10:21] Checking input data for scenario #0 ...
[INFO 10:21] Successfully validated all input files
[UI-event] Running scenario "testi_emme-matrices"
[INFO 10:21] Initializing Emme...
[INFO 10:21] Starting Emme...
[INFO 10:21] Emme started
[INFO 10:21] Initializing matrices and models...
[INFO 10:21] Starting simulation with 15 iterations...
[INFO 10:21] Calculates road charges for time period aht...
[INFO 10:21] Calculates road charges for time period pt...
[INFO 10:21] Calculates road charges for time period iht...
[INFO 10:21] Pedestrian assignment started...
[INFO 10:21] Pedestrian assignment performed for scenario 51
[INFO 10:21] Sets bike functions for scenario 51
[INFO 10:21] Bike assignment started...
[INFO 10:21] Bike assignment performed for scenario 51
[INFO 10:21] Sets car and transit functions for scenario 51
[INFO 10:21] Car assignment started...
[INFO 10:21] Car assignment performed for scenario 51
[INFO 10:21] Stopping criteria: RELATIVE_GAP, iteration 2 / 200
[INFO 10:21] Calculates cumulative travel times for scenario 51
[INFO 10:21] Transit assignment started...
[INFO 10:21] Transit assignment performed for scenario 51
[EXCEPTION] Traceback (most recent call last):
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\helmet.py", line 269, in <module>
[EXCEPTION]     main(args)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\helmet.py", line 96, in main
[EXCEPTION]     args.use_fixed_transit_cost, iterations==0)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\modelsystem.py", line 179, in assign_base_demand
[EXCEPTION]     self.ass_model.init_assign(base_demand)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\assignment\emme_assignment.py", line 104, in init_assign
[EXCEPTION]     self._copy_matrix("time", "bike", ap0, ap)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\assignment\emme_assignment.py", line 277, in _copy_matrix
[EXCEPTION]     "{} {}".format(to_mtx["description"], ass_period_2.name))
[EXCEPTION] TypeError: string indices must be integers
[ERROR] [...] TypeError: string indices must be integers
[UI-event] Scenario "testi_emme-matrices" finished

My setting were following:

2023-08-17 10:21:41 [DEBUG] helmet_version=v4.1.3
2023-08-17 10:21:41 [DEBUG] sys.version_info=3
2023-08-17 10:21:41 [DEBUG] sys.path=['C:\\Users\\PiippJo\\Downloads\\helmet-model-system-refactor-emme_matrices\\Scripts', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\python37.zip', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\DLLs', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37', 'C:\\Users\\PiippJo\\AppData\\Roaming\\Python\\Python37\\site-packages', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages\\win32', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages\\win32\\lib', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages\\Pythonwin']
2023-08-17 10:21:41 [DEBUG] log_level=DEBUG
2023-08-17 10:21:41 [DEBUG] log_format=JSON
2023-08-17 10:21:41 [DEBUG] end_assignment_only=False
2023-08-17 10:21:41 [DEBUG] is_agent_model=False
2023-08-17 10:21:41 [DEBUG] do_not_use_emme=False
2023-08-17 10:21:41 [DEBUG] separate_emme_scenarios=False
2023-08-17 10:21:41 [DEBUG] save_matrices=True
2023-08-17 10:21:41 [DEBUG] del_strat_files=True
2023-08-17 10:21:41 [DEBUG] scenario_name=testi_emme-matrices
2023-08-17 10:21:41 [DEBUG] results_path=E:\Results\Jopi_testi
2023-08-17 10:21:41 [DEBUG] emme_path=D:\helmet_41\Projektit\pienitestiprojekti\pienitestiprojekti.emp
2023-08-17 10:21:41 [DEBUG] first_scenario_id=51
2023-08-17 10:21:41 [DEBUG] first_matrix_id=400
2023-08-17 10:21:41 [DEBUG] baseline_data_path=C:\Users\PiippJo\Downloads\helmet-model-system-olusanya\Scripts\tests\test_data\Base_input_data
2023-08-17 10:21:41 [DEBUG] forecast_data_path=C:\Users\PiippJo\Downloads\helmet-model-system-olusanya\Scripts\tests\test_data\Scenario_input_data\2030_test
2023-08-17 10:21:41 [DEBUG] iterations=15
2023-08-17 10:21:41 [DEBUG] max_gap=1.0
2023-08-17 10:21:41 [DEBUG] rel_gap=0.01
2023-08-17 10:21:41 [DEBUG] use_fixed_transit_cost=False

zptro · 2023-08-17T07:59:00Z

Hi and thank you for your PR! I am sad to see the Emme matrix cheatsheet go but I understand the reasoning behind. 😊

I tried running this branch with the samll test network. I encountered an type error, maybe you can check it?

[UI-event] Initializing run of scenarios: testi_emme-matrices
[INFO 10:21] Checking base inputdata...
[INFO 10:21] Checking base zonedata & scenario-based input data...
[INFO 10:21] Checking input data for scenario #0 ...
[INFO 10:21] Successfully validated all input files
[UI-event] Running scenario "testi_emme-matrices"
[INFO 10:21] Initializing Emme...
[INFO 10:21] Starting Emme...
[INFO 10:21] Emme started
[INFO 10:21] Initializing matrices and models...
[INFO 10:21] Starting simulation with 15 iterations...
[INFO 10:21] Calculates road charges for time period aht...
[INFO 10:21] Calculates road charges for time period pt...
[INFO 10:21] Calculates road charges for time period iht...
[INFO 10:21] Pedestrian assignment started...
[INFO 10:21] Pedestrian assignment performed for scenario 51
[INFO 10:21] Sets bike functions for scenario 51
[INFO 10:21] Bike assignment started...
[INFO 10:21] Bike assignment performed for scenario 51
[INFO 10:21] Sets car and transit functions for scenario 51
[INFO 10:21] Car assignment started...
[INFO 10:21] Car assignment performed for scenario 51
[INFO 10:21] Stopping criteria: RELATIVE_GAP, iteration 2 / 200
[INFO 10:21] Calculates cumulative travel times for scenario 51
[INFO 10:21] Transit assignment started...
[INFO 10:21] Transit assignment performed for scenario 51
[EXCEPTION] Traceback (most recent call last):
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\helmet.py", line 269, in <module>
[EXCEPTION]     main(args)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\helmet.py", line 96, in main
[EXCEPTION]     args.use_fixed_transit_cost, iterations==0)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\modelsystem.py", line 179, in assign_base_demand
[EXCEPTION]     self.ass_model.init_assign(base_demand)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\assignment\emme_assignment.py", line 104, in init_assign
[EXCEPTION]     self._copy_matrix("time", "bike", ap0, ap)
[EXCEPTION]   File "C:\Users\PiippJo\Downloads\helmet-model-system-refactor-emme_matrices\Scripts\assignment\emme_assignment.py", line 277, in _copy_matrix
[EXCEPTION]     "{} {}".format(to_mtx["description"], ass_period_2.name))
[EXCEPTION] TypeError: string indices must be integers
[ERROR] [...] TypeError: string indices must be integers
[UI-event] Scenario "testi_emme-matrices" finished

My setting were following:

2023-08-17 10:21:41 [DEBUG] helmet_version=v4.1.3
2023-08-17 10:21:41 [DEBUG] sys.version_info=3
2023-08-17 10:21:41 [DEBUG] sys.path=['C:\\Users\\PiippJo\\Downloads\\helmet-model-system-refactor-emme_matrices\\Scripts', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\python37.zip', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\DLLs', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37', 'C:\\Users\\PiippJo\\AppData\\Roaming\\Python\\Python37\\site-packages', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages\\win32', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages\\win32\\lib', 'C:\\Program Files\\INRO\\Emme\\Emme 4\\Emme-23.00.01.23\\Python37\\lib\\site-packages\\Pythonwin']
2023-08-17 10:21:41 [DEBUG] log_level=DEBUG
2023-08-17 10:21:41 [DEBUG] log_format=JSON
2023-08-17 10:21:41 [DEBUG] end_assignment_only=False
2023-08-17 10:21:41 [DEBUG] is_agent_model=False
2023-08-17 10:21:41 [DEBUG] do_not_use_emme=False
2023-08-17 10:21:41 [DEBUG] separate_emme_scenarios=False
2023-08-17 10:21:41 [DEBUG] save_matrices=True
2023-08-17 10:21:41 [DEBUG] del_strat_files=True
2023-08-17 10:21:41 [DEBUG] scenario_name=testi_emme-matrices
2023-08-17 10:21:41 [DEBUG] results_path=E:\Results\Jopi_testi
2023-08-17 10:21:41 [DEBUG] emme_path=D:\helmet_41\Projektit\pienitestiprojekti\pienitestiprojekti.emp
2023-08-17 10:21:41 [DEBUG] first_scenario_id=51
2023-08-17 10:21:41 [DEBUG] first_matrix_id=400
2023-08-17 10:21:41 [DEBUG] baseline_data_path=C:\Users\PiippJo\Downloads\helmet-model-system-olusanya\Scripts\tests\test_data\Base_input_data
2023-08-17 10:21:41 [DEBUG] forecast_data_path=C:\Users\PiippJo\Downloads\helmet-model-system-olusanya\Scripts\tests\test_data\Scenario_input_data\2030_test
2023-08-17 10:21:41 [DEBUG] iterations=15
2023-08-17 10:21:41 [DEBUG] max_gap=1.0
2023-08-17 10:21:41 [DEBUG] rel_gap=0.01
2023-08-17 10:21:41 [DEBUG] use_fixed_transit_cost=False

Thank you both for testing this thoroughly (i.e., doing the job that I should have done). ❤️

johpiip

Test run succeeded and old and new matrices have same names, descriptions, and order. Nice work! 😊 Sorry for the delay in reviewing!

johpiip · 2023-08-17T10:43:45Z

Hey, sorry, I still have a question! By mistake, I ran the second model run with wrong scripts so even though everything work now, matrices are in different order and descriptions are different.

Other matrices are in same order are previously but trip_part matrices are differently. Does this clarify:

Is this preferred behaviour or should we have them in the same order?

Descriptions in this PR are more condensed than previously and do not add much new information on top of the matrix name. Did you consider having the same description info as earlier? Examples can be seen in the image above, too.

zptro · 2023-08-18T13:14:23Z

Hey, sorry, I still have a question! By mistake, I ran the second model run with wrong scripts so even though everything work now, matrices are in different order and descriptions are different.

Other matrices are in same order are previously but trip_part matrices are differently. Does this clarify:

Is this preferred behaviour or should we have them in the same order?

Descriptions in this PR are more condensed than previously and do not add much new information on top of the matrix name. Did you consider having the same description info as earlier? Examples can be seen in the image above, too.

Good comments!

The so-called trip-part matrix dictionary now goes directly into the specification of matrix results calculation (which means they are split into total matrices and matrices by mode subset), so that is the reason why it cannot be in exactly the same order as before. But I changed the order a bit so that it is as close as possible to what it was before.

I also changed the description to a more descriptive one, which is in fact also what it is in the Modeller tool specification.

hsl-petrhaj

Works on my computer. Tiny (about 1 ppm) differences in the results compared to the latest olusanya, not a problem by my standards.
The demand matrix looks the same, amount of trips is the same as well. I did not check the order of the matrices.

zptro added 3 commits June 20, 2023 11:45

Make handling of EMME matrices less error-prone and more flexible

ff8ac2e

Refactor matrix creation

312ab09

Also add docstrings and remove deprecated imports

Fix transit trip part matrix ids and expand matrix id space when needed

d8fc46b

zptro requested review from johpiip and hsl-petrhaj June 22, 2023 05:50

Add matrix name and description length check

7c527f0

hsl-petrhaj previously approved these changes Jun 29, 2023

View reviewed changes

hsl-petrhaj self-requested a review June 29, 2023 07:49

Fix reference

c90a919

zptro dismissed hsl-petrhaj’s stale review via c90a919 June 29, 2023 09:41

hsl-petrhaj previously approved these changes Jun 29, 2023

View reviewed changes

Remove extra attribute @walk

e67fdfa

zptro dismissed hsl-petrhaj’s stale review via e67fdfa June 29, 2023 12:02

zptro requested a review from hsl-petrhaj June 29, 2023 12:02

hsl-petrhaj added 2 commits June 29, 2023 15:04

Merge branch 'olusanya' into refactor/emme_matrices

c3efedd

Merge branch 'olusanya' into refactor/emme_matrices

df476e6

hsl-petrhaj requested changes Jun 29, 2023

View reviewed changes

Fix missing transit time matrix problem

108c251

johpiip requested changes Aug 17, 2023

View reviewed changes

Fix description in matrix copying

43b598f

zptro requested review from johpiip and hsl-petrhaj August 17, 2023 07:56

johpiip previously approved these changes Aug 17, 2023

View reviewed changes

Change matrix description for transit impedance matrices

b017811

zptro dismissed johpiip’s stale review via b017811 August 18, 2023 13:02

Merge branch 'olusanya' into refactor/emme_matrices

9575abb

zptro requested a review from johpiip August 22, 2023 09:04

Merge branch 'olusanya' into refactor/emme_matrices

3876600

hsl-petrhaj approved these changes Sep 1, 2023

View reviewed changes

zptro merged commit 5a14cc7 into olusanya Sep 4, 2023
2 checks passed

zptro deleted the refactor/emme_matrices branch September 4, 2023 05:43

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Make handling of EMME matrices less error-prone and more flexible #525

Make handling of EMME matrices less error-prone and more flexible #525

zptro commented Jun 20, 2023 •

edited

Loading

zptro commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj left a comment

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023 •

edited

Loading

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj left a comment

zptro commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023 •

edited

Loading

hsl-petrhaj left a comment

zptro commented Aug 1, 2023

zptro commented Aug 1, 2023

johpiip left a comment

zptro commented Aug 17, 2023

johpiip left a comment

johpiip commented Aug 17, 2023

zptro commented Aug 18, 2023

hsl-petrhaj left a comment

	dict
	key : str
	Assignment class (car_work/transit_leisure/...)
	value : dict
	key : str
	Matrix type (demand/time/cost/dist/...)
	value : str
	EMME matrix id

Make handling of EMME matrices less error-prone and more flexible #525

Make handling of EMME matrices less error-prone and more flexible #525

Conversation

zptro commented Jun 20, 2023 • edited Loading

zptro commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj left a comment

Choose a reason for hiding this comment

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023 • edited Loading

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj left a comment

Choose a reason for hiding this comment

zptro commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023

hsl-petrhaj commented Jun 29, 2023 • edited Loading

hsl-petrhaj left a comment

Choose a reason for hiding this comment

zptro commented Aug 1, 2023

zptro commented Aug 1, 2023

johpiip left a comment

Choose a reason for hiding this comment

zptro commented Aug 17, 2023

johpiip left a comment

Choose a reason for hiding this comment

johpiip commented Aug 17, 2023

zptro commented Aug 18, 2023

hsl-petrhaj left a comment

Choose a reason for hiding this comment

zptro commented Jun 20, 2023 •

edited

Loading

hsl-petrhaj commented Jun 29, 2023 •

edited

Loading

hsl-petrhaj commented Jun 29, 2023 •

edited

Loading