Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Rhorsey/sampling v2 plotting patch #217

Merged
merged 4 commits into from
Sep 23, 2024
Merged

Rhorsey/sampling v2 plotting patch #217

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
6f68102
Updating plotting scripts to sampling V2. AMI todo and EIA needs a qu…
rHorsey Sep 18, 2024
79d6ec1
Adding license to apportionment.
rHorsey Sep 18, 2024
d35ae57
Updating downloading of comstock data from s3.
rHorsey Sep 19, 2024
f7faead
Addressing Chriss comments - many thanks!
rHorsey Sep 19, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 12 additions & 11 deletions postprocessing/compare_comstock_to_cbecs.py.template
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,12 +32,6 @@ def main():
stock_estimation_version='2024R2', # Only updated when a new stock estimate is published
truth_data_version='v01' # Typically don't change this
)

# Scale ComStock run to CBECS 2018 AND remove non-ComStock buildings from CBECS
comstock.add_weights_aportioned_by_stock_estimate(apportionment=stock_estimate)
comstock.create_national_aggregation()
comstock.create_geospatially_resolved_aggregations(comstock.STATE_ID, pretty_geo_col_name='state_id')
comstock.create_geospatially_resolved_aggregations(comstock.COUNTY_ID, pretty_geo_col_name='county_id')

# CBECS
cbecs = cspp.CBECS(
Expand All @@ -46,19 +40,26 @@ def main():
color_hex='#009E73', # Color used to represent CBECS in plots
reload_from_csv=False # True if CSV already made and want faster reload times
)
# TODO Update past here including ensuring we can still apply CBECS weights on top of previous weights.


# Scale ComStock runs to the 'truth data' from StockE V3 estimates using bucket-based apportionment
comstock.add_weights_aportioned_by_stock_estimate(apportionment=stock_estimate)
# Scale ComStock run to CBECS 2018 AND remove non-ComStock buildings from CBECS
comstock.add_national_scaling_weights(cbecs, remove_non_comstock_bldg_types_from_cbecs=True)
comstock.calculate_weighted_columnal_values()
comstock.export_to_csv_wide()
# TODO This needs to be rewritten with safe column names, lazyframe usage, etc.
#comstock.calculate_weighted_columnal_values()

# Uncomment whichever to write results to disk:
comstock.create_national_aggregation()
# comstock.create_geospatially_resolved_aggregations(comstock.STATE_ID, pretty_geo_col_name='state_id')
# comstock.create_geospatially_resolved_aggregations(comstock.COUNTY_ID, pretty_geo_col_name='county_id')

# Make a comparison by passing in a list of CBECs and ComStock runs to compare
# upgrade_id can be 'All' or the upgrade number
comstock.create_plotting_lazyframe()
comp = cspp.ComStockToCBECSComparison(cbecs_list=[cbecs], comstock_list=[comstock], upgrade_id='All',make_comparison_plots=True)

comp.export_to_csv_wide()


# Code to execute the script
if __name__ == "__main__":
Expand Down
42 changes: 24 additions & 18 deletions postprocessing/compare_runs.py.template
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,14 +12,14 @@ logger = logging.getLogger(__name__)
def main():
# First ComStock run
comstock_a = cspp.ComStock(
s3_base_dir='eulp/comstock_fy22', # If run not on S3, download results_up**.parquet manually
comstock_run_name='com_v15_cooking', # Name of the run on S3
comstock_run_version='v15', # Use whatever you want to see in plot and folder names
s3_base_dir='eulp/euss_com', # If run not on S3, download results_up**.parquet manually
comstock_run_name='sampling_lighting_11079_1', # Name of the run on S3
comstock_run_version='sampling_lighting_11079_1', # Use whatever you want to see in plot and folder names
comstock_year=2018, # Typically don't change this
athena_table_name=None, # Typically don't change this
truth_data_version='v01', # Typically don't change this
buildstock_csv_name='buildstock.csv', # Download buildstock.csv manually
acceptable_failure_percentage=0.05, # Can increase this when testing and high failure are OK
acceptable_failure_percentage=0.25, # Can increase this when testing and high failure are OK
drop_failed_runs=True, # False if you want to evaluate which runs failed in raw output data
color_hex='#0072B2', # Color used to represent this run in plots
skip_missing_columns=True, # False if you want to ensure you have all data specified for export
Expand All @@ -29,21 +29,27 @@ def main():

# Second ComStock run
comstock_b = cspp.ComStock(
s3_base_dir='eulp/comstock_fy22', # If run not on S3, download results_up**.parquet manually
comstock_run_name='com_v16_windows_lighting', # Name of the run on S3
comstock_run_version='v16', # Use whatever you want to see in plot and folder names
s3_base_dir='eulp/euss_com', # If run not on S3, download results_up**.parquet manually
comstock_run_name='cycle_4_sampling_test_rand_985932_20240321', # Name of the run on S3
comstock_run_version='new_sampling_test', # Use whatever you want to see in plot and folder names
comstock_year=2018, # Typically don't change this
athena_table_name=None, # Typically don't change this
athena_table_name='rand_985932_20240321', # Typically same as comstock_run_name or None
truth_data_version='v01', # Typically don't change this
buildstock_csv_name='buildstock.csv', # Download buildstock.csv manually
acceptable_failure_percentage=0.05, # Can increase this when testing and high failure are OK
buildstock_csv_name='rand_985932_sampling_buildstock.csv', # Download buildstock.csv manually
acceptable_failure_percentage=0.9, # Can increase this when testing and high failure are OK
drop_failed_runs=True, # False if you want to evaluate which runs failed in raw output data
color_hex='#56B4E9', # Color used to represent this run in plots
skip_missing_columns=True, # False if you want to ensure you have all data specified for export
reload_from_csv=False, # True if CSV already made and want faster reload times
include_upgrades=False # False if not looking at upgrades
)

# Stock Estimation for Apportionment:
stock_estimate = cspp.Apportion(
stock_estimation_version='2024R2', # Only updated when a new stock estimate is published
truth_data_version='v01' # Typically don't change this
)

# CBECS
cbecs = cspp.CBECS(
cbecs_year=2018, # 2012 and 2018 currently available
Expand All @@ -52,19 +58,19 @@ def main():
reload_from_csv=False # True if CSV already made and want faster reload times
)

# Scale both ComStock runs to CBECS 2018 AND remove non-ComStock buildings from CBECS
# First scale ComStock runs to the 'truth data' from StockE V3 estimates using bucket-based apportionment
# Then scale both ComStock runs to CBECS 2018 AND remove non-ComStock buildings from CBECS
# This is how weights in the models are set to represent national energy consumption
comstock_a.add_weights_aportioned_by_stock_estimate(apportionment=stock_estimate)
comstock_a.add_national_scaling_weights(cbecs, remove_non_comstock_bldg_types_from_cbecs=True)
comstock_b.add_weights_aportioned_by_stock_estimate(apportionment=stock_estimate)
comstock_b.add_national_scaling_weights(cbecs, remove_non_comstock_bldg_types_from_cbecs=True)

# Uncomment this to correct gas consumption for a ComStock run to match CBECS
# Don't typically want to do this
# comstock_a.correct_comstock_gas_to_match_cbecs(cbecs)

# Export CBECS and ComStock data to wide and long formats for Tableau and to skip processing later
cbecs.export_to_csv_wide() # May comment this out if CSV output isn't needed
comstock_a.export_to_csv_wide() # May comment this out if CSV output isn't needed
comstock_b.export_to_csv_wide() # May comment this out if CSV output isn't needed
# comstock_a.create_national_aggregation() # May comment this out if CSV output isn't needed
# comstock_b.create_national_aggregation() # May comment this out if CSV output isn't needed
# TODO This (long CSV export) is not yet re-implemented
# comstock_a.export_to_csv_long() # Long format useful for stacking end uses and fuels
# comstock_b.export_to_csv_long() # Long format useful for stacking end uses and fuels

Expand All @@ -73,7 +79,7 @@ def main():
cbecs_list=[cbecs],
comstock_list = [comstock_a, comstock_b],
make_comparison_plots=True
)
)

# Export the comparison data to wide format for Tableau
comparison.export_to_csv_wide()
Expand Down
136 changes: 73 additions & 63 deletions postprocessing/compare_upgrades.py.template
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,63 +1,73 @@
#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import logging

import comstockpostproc as cspp


logging.basicConfig(level='INFO') # Use DEBUG, INFO, or WARNING
logger = logging.getLogger(__name__)

def main():
# ComStock run
comstock = cspp.ComStock(
s3_base_dir='eulp/euss_com', # If run not on S3, download results_up**.parquet manually
comstock_run_name='hprtu_stdperf_fan_test_10k', # Name of the run on S3
comstock_run_version='hprtu_stdperf_fan_test_10k', # Use whatever you want to see in plot and folder names
comstock_year=2018, # Typically don't change this
athena_table_name=None, # Typically don't change this
truth_data_version='v01', # Typically don't change this
buildstock_csv_name='buildstock.csv', # Download buildstock.csv manually
acceptable_failure_percentage=0.025, # Can increase this when testing and high failure are OK
drop_failed_runs=True, # False if you want to evaluate which runs failed in raw output data
color_hex='#0072B2', # Color used to represent this run in plots
skip_missing_columns=True, # False if you want to ensure you have all data specified for export
reload_from_csv=False, # True if CSV already made and want faster reload times
include_upgrades=True, # False if not looking at upgrades
upgrade_ids_to_skip=[], # Use [1, 3] etc. to exclude certain upgrades
make_timeseries_plots=True,
states={
#'MN': 'Minnesota', # specify state to use for timeseries plots in dictionary format. State ID must correspond correctly.
'MA':'Massachusetts',
'OR': 'Oregon',
'LA': 'Louisiana',
#'AZ': 'Arizona',
#'TN': 'Tennessee'
},
upgrade_ids_for_comparison={} # Use {'<Name you want for comparison run folder>':[0,1,2]}; add as many upgrade IDs as needed, but plots look strange over 5
)

# CBECS
cbecs = cspp.CBECS(
cbecs_year=2018, # 2012 and 2018 currently available
truth_data_version='v01', # Typically don't change this
color_hex='#009E73', # Color used to represent CBECS in plots
reload_from_csv=False # True if CSV already made and want faster reload times
)

# Scale ComStock run to CBECS 2018 AND remove non-ComStock buildings from CBECS
# This is how weights in the models are set to represent national energy consumption
comstock.add_national_scaling_weights(cbecs, remove_non_comstock_bldg_types_from_cbecs=True)

# Export CBECS and ComStock data to wide and long formats for Tableau and to skip processing later
cbecs.export_to_csv_wide() # May comment this out after run once
comstock.export_to_csv_wide() # May comment this out after run once
# comstock.export_to_csv_long() # Long format useful for stacking end uses and fuels

# Create measure run comparisons; only use if run has measures
comparison = cspp.ComStockMeasureComparison(comstock, states=comstock.states, make_comparison_plots = comstock.make_comparison_plots, make_timeseries_plots = comstock.make_timeseries_plots)

# Code to execute the script
if __name__=="__main__":
main()
#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import logging

import comstockpostproc as cspp


logging.basicConfig(level='INFO') # Use DEBUG, INFO, or WARNING
logger = logging.getLogger(__name__)

def main():
# ComStock run
comstock = cspp.ComStock(
s3_base_dir='eulp/euss_com', # If run not on S3, download results_up**.parquet manually
comstock_run_name='sampling_lighting_11079_1', # Name of the run on S3
comstock_run_version='sampling_lighting_11079_1', # Use whatever you want to see in plot and folder names
comstock_year=2018, # Typically don't change this
athena_table_name=None, # Typically don't change this
truth_data_version='v01', # Typically don't change this
buildstock_csv_name='buildstock.csv', # Download buildstock.csv manually
acceptable_failure_percentage=0.25, # Can increase this when testing and high failure are OK
drop_failed_runs=True, # False if you want to evaluate which runs failed in raw output data
color_hex='#0072B2', # Color used to represent this run in plots
skip_missing_columns=True, # False if you want to ensure you have all data specified for export
reload_from_csv=False, # True if CSV already made and want faster reload times
include_upgrades=True, # False if not looking at upgrades
upgrade_ids_to_skip=[], # Use [1, 3] etc. to exclude certain upgrades
make_timeseries_plots=False,
states={
#'MN': 'Minnesota', # specify state to use for timeseries plots in dictionary format. State ID must correspond correctly.
'MA':'Massachusetts',
#'OR': 'Oregon',
#'LA': 'Louisiana',
#'AZ': 'Arizona',
#'TN': 'Tennessee'
},
upgrade_ids_for_comparison={} # Use {'<Name you want for comparison run folder>':[0,1,2]}; add as many upgrade IDs as needed, but plots look strange over 5
)

# Stock Estimation for Apportionment:
stock_estimate = cspp.Apportion(
stock_estimation_version='2024R2', # Only updated when a new stock estimate is published
truth_data_version='v01' # Typically don't change this
)

# CBECS
cbecs = cspp.CBECS(
cbecs_year=2018, # 2012 and 2018 currently available
truth_data_version='v01', # Typically don't change this
color_hex='#009E73', # Color used to represent CBECS in plots
reload_from_csv=False # True if CSV already made and want faster reload times
)

# Scale ComStock runs to the 'truth data' from StockE V3 estimates using bucket-based apportionment
comstock.add_weights_aportioned_by_stock_estimate(apportionment=stock_estimate)
# Scale ComStock run to CBECS 2018 AND remove non-ComStock buildings from CBECS
comstock.add_national_scaling_weights(cbecs, remove_non_comstock_bldg_types_from_cbecs=True)

# Export CBECS and ComStock data to wide and long formats for Tableau and to skip processing later
# cbecs.export_to_csv_wide() # May comment this out after run once
# comstock.create_national_aggregation()
# comstock.create_geospatially_resolved_aggregations(comstock.STATE_ID, pretty_geo_col_name='state_id')
# comstock.create_geospatially_resolved_aggregations(comstock.COUNTY_ID, pretty_geo_col_name='county_id')
# TODO Long is def not working as expected anymore...
# comstock.export_to_csv_long() # Long format useful for stacking end uses and fuels
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Where is ComStock wide export?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That's the create_national_aggregations method now - same code under the hood but redone to get correctly weighted national level results!


# Create measure run comparisons; only use if run has measures
comparison = cspp.ComStockMeasureComparison(comstock, states=comstock.states, make_comparison_plots = comstock.make_comparison_plots, make_timeseries_plots = comstock.make_timeseries_plots)

# Code to execute the script
if __name__=="__main__":
main()
7 changes: 6 additions & 1 deletion postprocessing/comstockpostproc/cbecs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -483,4 +483,9 @@ def export_to_csv_wide(self):

file_name = f'CBECS wide.csv'
file_path = os.path.join(self.output_dir, file_name)
self.data.to_csv(file_path, index=False)
try:
self.data.sink_csv(file_path)
except pl.exceptions.InvalidOperationError:
logger.warn('Warning - sink_csv not supported for metadata write in current polars version')
logger.warn('Falling back to .collect.write_csv')
self.data.collect().write_csv(file_path)
Loading