Sampler and Workflow Refactor

aws_demo_project.yml

@@ -1,33 +1,36 @@
-schema_version: 0.2
-stock_type: residential
+schema_version: '0.3'
 buildstock_directory: ../resstock  # Relative to this file or absolute
 project_directory: project_national  # Relative to buildstock_directory
 output_directory: ../demo_test_outputs
 weather_files_url: https://data.nrel.gov/system/files/128/ResStock_TMY3.zip
 
 baseline:
-  n_datapoints: 4  # Comment this line out if using a custom buildstock csv file
   n_buildings_represented: 133172057  # Total number of residential dwelling units in contiguous United States, including unoccupied units, resulting from acensus tract level query of ACS 5-yr 2016 (i.e. 2012-2016), using this script: https://github.com/NREL/resstock-estimation/blob/master/sources/spatial/tsv_maker.py.
-  sampling_algorithm: quota # The default resstock sampling algorithm - use precomputed if using the precomputed_sample option
+
+sampler:
+  type: residential_quota # change to residential_quota_downselect to do downselect
+  args:
+    n_datapoints: 4
+    # resample: false # Uncomment and set specify logic you you want to downselect to a subset of the building stock
+    # logic:
+    #   - Geometry Building Type RECS|Single-Family Detached
+    #   - Vacancy Status|Occupied
+
+workflow_generator:
+  type: residential_default  
+  args:
+    timeseries_csv_export:
+      reporting_frequency: Hourly
+      include_enduse_subcategories: true
 
 upgrades:
   - upgrade_name: Triple-Pane Windows
     options:
       - option: Windows|Low-E, Triple, Non-metal, Air, L-Gain
-#        apply_logic:
         costs:
           - value: 45.77
             multiplier: Window Area (ft^2)
         lifetime: 30
-timeseries_csv_export:
-  reporting_frequency: Hourly
-  include_enduse_subcategories: true
-
-# downselect: # Uncomment and set specify logic you you want to downselect to a subset of the building stock
-#   resample: true
-#   logic:
-#     - Geometry Building Type RECS|Single-Family Detached
-#     - Vacancy Status|Occupied
 
 aws:
   # The job_identifier must be unique, start with alpha, not include dashes, and limited to 10 chars

buildstockbatch/__version__.py

@@ -1,9 +1,12 @@
+import datetime as dt
+
+
 __title__ = 'buildstockbatch'
 __description__ = 'Executing BuildStock projects on batch infrastructure.'
 __url__ = 'http://github.com/NREL/buildstockbatch'
-__version__ = '0.19'
-__schema_version__ = '0.2'
+__version__ = '0.20'
+__schema_version__ = '0.3'
 __author__ = 'Noel Merket'
 __author_email__ = 'noel.merket@nrel.gov'
 __license__ = 'BSD-3'
-__copyright__ = 'Copyright 2020 The Alliance for Sustainable Energy'
+__copyright__ = 'Copyright {} The Alliance for Sustainable Energy'.format(dt.date.today().year)

buildstockbatch/aws/aws.py

@@ -40,7 +40,7 @@
 from buildstockbatch.base import ValidationError
 from buildstockbatch.aws.awsbase import AwsJobBase
 from buildstockbatch import postprocessing
-from ..utils import log_error_details
+from ..utils import log_error_details, get_project_configuration
 
 logger = logging.getLogger(__name__)
 
@@ -1661,7 +1661,7 @@ def __init__(self, project_filename):
 
     @staticmethod
     def validate_instance_types(project_file):
-        cfg = AwsBatch.get_project_configuration(project_file)
+        cfg = get_project_configuration(project_file)
         aws_config = cfg['aws']
         boto3_session = boto3.Session(region_name=aws_config['region'])
         ec2 = boto3_session.client('ec2')
@@ -1772,10 +1772,7 @@ def run_batch(self):
         """
 
         # Generate buildstock.csv
-        if 'downselect' in self.cfg:
-            buildstock_csv_filename = self.downselect()
-        else:
-            buildstock_csv_filename = self.run_sampling()
+        buildstock_csv_filename = self.sampler.run_sampling()
 
         # Compress and upload assets to S3
         with tempfile.TemporaryDirectory() as tmpdir, tempfile.TemporaryDirectory() as tmp_weather_dir:
@@ -1837,6 +1834,7 @@ def run_batch(self):
                 with open(job_json_filename, 'w') as f:
                     json.dump({
                         'job_num': i,
+                        'n_datapoints': n_datapoints,
                         'batch': batch,
                     }, f, indent=4)
             array_size = i
@@ -1981,7 +1979,7 @@ def run_job(cls, job_id, bucket, prefix, job_name, region):
                 sim_id = f'bldg{building_id:07d}up{upgrade_id:02d}'
 
                 # Create OSW
-                osw = cls.create_osw(cfg, sim_id, building_id, upgrade_idx)
+                osw = cls.create_osw(cfg, jobs_d['n_datapoints'], sim_id, building_id, upgrade_idx)
                 with open(os.path.join(sim_dir, 'in.osw'), 'w') as f:
                     json.dump(osw, f, indent=4)