Original error message, thrown when using the national wind blade dataset to run 1-2 years ending at 2016:
60587 - 2016 - Transportation - transportation shortcut calculations done
LCA inventory does not exist for 2016 in use glass fiber
LCA emissions inventory does not exist for 2016 in use glass fiber
LCA inventory does not exist for 2016 in use epoxy
LCA emissions inventory does not exist for 2016 in use epoxy
60672 - 2016 - Transportation - transportation shortcut calculations done
Creating diagnostic visualizations for run 0 at 3589 s
Traceback (most recent call last):
File "/Users/mayadl/anaconda3/envs/celavi/lib/python3.8/runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/Users/mayadl/anaconda3/envs/celavi/lib/python3.8/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/Users/mayadl/Documents/GitHub/celavi/celavi/__main__.py", line 25, in <module>
Scenario(parser=PARSER)
File "/Users/mayadl/Documents/GitHub/celavi/celavi/scenario.py", line 119, in __init__
self.postprocess()
File "/Users/mayadl/Documents/GitHub/celavi/celavi/scenario.py", line 543, in postprocess
self.netw.save_costgraph_outputs()
File "/Users/mayadl/Documents/GitHub/celavi/celavi/costgraph.py", line 1061, in save_costgraph_outputs
_out = pd.DataFrame(
File "/Users/mayadl/anaconda3/envs/celavi/lib/python3.8/site-packages/pandas/core/frame.py", line 8884, in explode
counts0 = self[columns[0]].apply(mylen)
File "/Users/mayadl/anaconda3/envs/celavi/lib/python3.8/site-packages/pandas/core/frame.py", line 3761, in __getitem__
indexer = self.columns.get_loc(key)
File "/Users/mayadl/anaconda3/envs/celavi/lib/python3.8/site-packages/pandas/core/indexes/range.py", line 349, in get_loc
raise KeyError(key)
KeyError: 'destination_facility_id'
Same error message observed when running the national dataset for 2020 and for 2010-2012.
Bug was reproduced using the tiny dataset and setting the start and end years to various 2-3 year time spans around 2016, and was traced back to an empty pathway_crit_history in CostGraph. pathway_crit_history was empty because it makes a record of EOL pathways selected according to the criterion ("crit") value, and no wind blades had entered EOL due to the brief time span. During postprocessing, pathway_crit_history is munged into a human readable format and saved to file as part of the simulation results. When it is empty, the munging and saving fails and produces the above error message. Simulation results generated and postprocessed before this code is executed are saved normally.
Resolution
This is technically not a bug because CELAVI is running as intended, but the error can be better handled to provide useful feedback to the user.
Lines 1061-1072 in costgraph.py will be inserted into a try/except statement that postprocesses normally (maintains current functionality) when there is no error thrown, and prints an error message when a KeyError is detected. This will allow the code to complete execution and print the "simulation finished" message, without saving the empty pathway criterion history.
rjhanes
commented
3 months ago
Bug description
Original error message, thrown when using the national wind blade dataset to run 1-2 years ending at 2016:
Same error message observed when running the national dataset for 2020 and for 2010-2012.
Bug was reproduced using the tiny dataset and setting the start and end years to various 2-3 year time spans around 2016, and was traced back to an empty
pathway_crit_historyinCostGraph.pathway_crit_historywas empty because it makes a record of EOL pathways selected according to the criterion ("crit") value, and no wind blades had entered EOL due to the brief time span. During postprocessing,pathway_crit_historyis munged into a human readable format and saved to file as part of the simulation results. When it is empty, the munging and saving fails and produces the above error message. Simulation results generated and postprocessed before this code is executed are saved normally.Resolution
This is technically not a bug because CELAVI is running as intended, but the error can be better handled to provide useful feedback to the user.
Lines 1061-1072 in costgraph.py will be inserted into a try/except statement that postprocesses normally (maintains current functionality) when there is no error thrown, and prints an error message when a KeyError is detected. This will allow the code to complete execution and print the "simulation finished" message, without saving the empty pathway criterion history.