Skip to content

Commit

Permalink
A broken commit
Browse files Browse the repository at this point in the history
  • Loading branch information
@wenyikuang
wenyikuang committed Jan 9, 2024
1 parent 5bdd5df commit 9328329
Show file tree
Hide file tree
Showing 2 changed files with 85 additions and 85 deletions.
2 changes: 1 addition & 1 deletion Jenkinsfile_main_develop
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,6 @@


// Build for PR to main branch only.
if ((env.CHANGE_ID) && (env.CHANGE_TARGET == "main") ) {
if ((env.CHANGE_ID) && (env.CHANGE_TARGET) ) {
building_comstock_all()
}
168 changes: 84 additions & 84 deletions sampling/tsv_resampling_wrapper.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -46,90 +46,90 @@ def attempt_sampling(df_in, tsv_version, sim_year, buildstock_dir, sizing_arg):
gen_county_samples(row, tsv_dir_short, tsv_dir, sim_year, buildstock_dir, sizing_arg)


def gen_county_samples(row, tsv_dir_short, tsv_dir, sim_year, buildstock_dir, sizing_arg):
"""
Run tsv_resampling_small_batch.py on a county-by-county basis for counties that require additional samples.\
For each county, copy the contents of the comstock_tsvs folder to temporary_folder for data manipulation.\
The county_id.tsv and climate_zone.tsv files are modified within temporary_folder. In county_id.tsv, all\
probabilities are set to 0 except where the column is the county in question and the climate zone is that of where\
the county is located. In climate_zone.tsv, all probabilities are set to 0 except the climate zone where the\
county is located.
These manipulated tsvs and the number of additional samples needed are passed to tsv_resampling_small_batch.py\
for county-specific sampling. The output buildstock.csv is uniquely named with the county ID and saved in a\
separate folder. temporary_folder is deleted, and the loop continues to the next county in need_more_samples.csv.
:param row: Specific row in file containing counties and number of additional samples
:param tsv_version: tsv version
:param tsv_dir: Directory where tsv files are located
:param sim_year: Simulation year (2012-2019)
:param buildstock_dir: Directory where county-specific buildstock.csv files are saved
"""
county = row['county_id']
print('county =', county)

# Extract TSVs to temporary folder
tmp_dir = tempfile.mkdtemp()
with zipfile.ZipFile(f'{tsv_dir}.zip', 'r') as zipObj:
zipObj.extractall(tmp_dir)

# Import year_of_simulation, county_id and climate_zone to manipulate
year_of_simulation = pd.read_csv(os.path.join(tmp_dir, tsv_dir_short, 'year_of_simulation.tsv'), sep='\t', index_col=False)
county_id = pd.read_csv(os.path.join(tmp_dir, tsv_dir_short, 'county_id.tsv'), sep='\t', index_col=False)
climate_zone = pd.read_csv(os.path.join(tmp_dir, tsv_dir_short, 'climate_zone.tsv'), sep='\t', index_col=False)

# Zero out all probabilities except for the simulation year - change probability to 1
for col in year_of_simulation.columns:
if col == 'Option=' + str(sim_year):
year_of_simulation[col].values[:] = 1
elif col != 'Option=' + str(sim_year):
year_of_simulation[col].values[:] = 0

# Zero out all probabilities except for the county in question - change probability to 1
# Note location of non-zero probability for identifying the county's climate zone
y = 0
for col in county_id.columns:
if col == 'Dependency=climate_zone':
continue
elif col == 'Option=' + row['county_id']:
y = int(county_id[county_id[col] != 0].index[0])
county_id.loc[y, col] = 1
elif col != 'Option=' + row['county_id']:
county_id[col].values[:] = 0

# Identify climate zone county is located in
zone = county_id.iloc[y, 0]

# Zero all climate zones except zone county is located in
# Change probability to 1
for col in climate_zone.columns:
if col == ('Option=' + zone):
climate_zone[col].values[:] = 1
elif col != ('Option=' + zone):
climate_zone[col].values[:] = 0

# Write altered TSVs to temporary folder for use in tsv_resampling
year_of_simulation.to_csv(os.path.join(tmp_dir, tsv_dir_short, 'year_of_simulation.tsv'), sep='\t', index=False)
county_id.to_csv(os.path.join(tmp_dir, tsv_dir_short, 'county_id.tsv'), sep='\t', index=False)
climate_zone.to_csv(os.path.join(tmp_dir, tsv_dir_short, 'climate_zone.tsv'), sep='\t', index=False)

# Number of additional samples needed for this county
n_add_samples = int(row['additional_samples'])
print('number of samples =', n_add_samples)

# Run sampling for county
lock_dir = tempfile.mkdtemp()
sampler = tsv_resampling_small_batch.instantiate_sampler(
buildstock_dir,
n_add_samples,
os.path.join(tmp_dir, tsv_dir_short),
lock_dir
)
sampler.run_sampling(n_datapoints=n_add_samples, county_id=county, sizing_arg=sizing_arg)
time.sleep(1)

# Remove temporary folder
shutil.rmtree(tmp_dir)
shutil.rmtree(lock_dir)
# def gen_county_samples(row, tsv_dir_short, tsv_dir, sim_year, buildstock_dir, sizing_arg):
# """
# Run tsv_resampling_small_batch.py on a county-by-county basis for counties that require additional samples.\
# For each county, copy the contents of the comstock_tsvs folder to temporary_folder for data manipulation.\
# The county_id.tsv and climate_zone.tsv files are modified within temporary_folder. In county_id.tsv, all\
# probabilities are set to 0 except where the column is the county in question and the climate zone is that of where\
# the county is located. In climate_zone.tsv, all probabilities are set to 0 except the climate zone where the\
# county is located.

# These manipulated tsvs and the number of additional samples needed are passed to tsv_resampling_small_batch.py\
# for county-specific sampling. The output buildstock.csv is uniquely named with the county ID and saved in a\
# separate folder. temporary_folder is deleted, and the loop continues to the next county in need_more_samples.csv.
# :param row: Specific row in file containing counties and number of additional samples
# :param tsv_version: tsv version
# :param tsv_dir: Directory where tsv files are located
# :param sim_year: Simulation year (2012-2019)
# :param buildstock_dir: Directory where county-specific buildstock.csv files are saved
# """
# county = row['county_id']
# print('county =', county)

# # Extract TSVs to temporary folder
# tmp_dir = tempfile.mkdtemp()
# with zipfile.ZipFile(f'{tsv_dir}.zip', 'r') as zipObj:
# zipObj.extractall(tmp_dir)

# # Import year_of_simulation, county_id and climate_zone to manipulate
# year_of_simulation = pd.read_csv(os.path.join(tmp_dir, tsv_dir_short, 'year_of_simulation.tsv'), sep='\t', index_col=False)
# county_id = pd.read_csv(os.path.join(tmp_dir, tsv_dir_short, 'county_id.tsv'), sep='\t', index_col=False)
# climate_zone = pd.read_csv(os.path.join(tmp_dir, tsv_dir_short, 'climate_zone.tsv'), sep='\t', index_col=False)

# # Zero out all probabilities except for the simulation year - change probability to 1
# for col in year_of_simulation.columns:
# if col == 'Option=' + str(sim_year):
# year_of_simulation[col].values[:] = 1
# elif col != 'Option=' + str(sim_year):
# year_of_simulation[col].values[:] = 0

# # Zero out all probabilities except for the county in question - change probability to 1
# # Note location of non-zero probability for identifying the county's climate zone
# y = 0
# for col in county_id.columns:
# if col == 'Dependency=climate_zone':
# continue
# elif col == 'Option=' + row['county_id']:
# y = int(county_id[county_id[col] != 0].index[0])
# county_id.loc[y, col] = 1
# elif col != 'Option=' + row['county_id']:
# county_id[col].values[:] = 0

# # Identify climate zone county is located in
# zone = county_id.iloc[y, 0]

# # Zero all climate zones except zone county is located in
# # Change probability to 1
# for col in climate_zone.columns:
# if col == ('Option=' + zone):
# climate_zone[col].values[:] = 1
# elif col != ('Option=' + zone):
# climate_zone[col].values[:] = 0

# # Write altered TSVs to temporary folder for use in tsv_resampling
# year_of_simulation.to_csv(os.path.join(tmp_dir, tsv_dir_short, 'year_of_simulation.tsv'), sep='\t', index=False)
# county_id.to_csv(os.path.join(tmp_dir, tsv_dir_short, 'county_id.tsv'), sep='\t', index=False)
# climate_zone.to_csv(os.path.join(tmp_dir, tsv_dir_short, 'climate_zone.tsv'), sep='\t', index=False)

# # Number of additional samples needed for this county
# n_add_samples = int(row['additional_samples'])
# print('number of samples =', n_add_samples)

# # Run sampling for county
# lock_dir = tempfile.mkdtemp()
# sampler = tsv_resampling_small_batch.instantiate_sampler(
# buildstock_dir,
# n_add_samples,
# os.path.join(tmp_dir, tsv_dir_short),
# lock_dir
# )
# sampler.run_sampling(n_datapoints=n_add_samples, county_id=county, sizing_arg=sizing_arg)
# time.sleep(1)

# # Remove temporary folder
# shutil.rmtree(tmp_dir)
# shutil.rmtree(lock_dir)


def parse_arguments():
Expand Down

0 comments on commit 9328329

Please sign in to comment.