Sampling regions characteristic

.github/workflows/ci.yml

@@ -28,7 +28,7 @@ jobs:
         with:
           repository: NatLabRockies/resstock
           path: resstock
-          ref: develop
+          ref: sampling_regions
       - name: Remove AWS from resstock yaml
         run: |
           cd resstock

buildstockbatch/local.py

@@ -114,6 +114,12 @@ def run_building(
             (resources_path / "hpxml-measures").symlink_to(hpxml_measures_path, target_is_directory=True)
         else:
             resources_path = None
+        # samplers_path = buildstock_path / "samplers"
+        # if samplers_path.exists():
+        # sampler_path = sim_path / "samplers"
+        # (sampler_path).symlink_to(samplers_path, target_is_directory=True)
+        # else:
+        # sampler_path = None
         custom_gems_path = buildstock_path / ".custom_gems"
         if custom_gems_path.exists():
             gems_path = sim_path / ".custom_gems"
@@ -205,6 +211,8 @@ def run_building(
                 if resources_path:
                     (resources_path / "hpxml-measures").unlink()
                     resources_path.rmdir()
+                # if sampler_path:
+                # sampler_path.unlink()
                 if gems_path:
                     gems_path.unlink()
 

buildstockbatch/sampler/__init__.py

@@ -4,5 +4,6 @@
     ResidentialQuotaSampler,
     ResidentialQuotaDownselectSampler,
 )  # noqa F041
+from .residential_stratified import ResidentialStratifiedSampler  # noqa F041
 from .commercial_sobol import CommercialSobolSampler  # noqa F041
 from .precomputed import PrecomputedSampler  # noqa F041

buildstockbatch/sampler/base.py

@@ -85,7 +85,7 @@ def run_sampling(self):
             return self._run_sampling_apptainer()
         else:
             assert self.container_runtime == ContainerRuntime.LOCAL_OPENSTUDIO
-            return self._run_sampling_local_openstudio()
+            return self._run_sampling_local()
 
     def _run_sampling_docker(self):
         """
@@ -103,9 +103,9 @@ def _run_sampling_apptainer(self):
         """
         raise NotImplementedError
 
-    def _run_sampling_local_openstudio(self):
+    def _run_sampling_local(self):
         """
-        Execute the sampling on the local openstudio instance
+        Execute the sampling on the local instance
 
         Replace this in a subclass as necessary
         """

buildstockbatch/sampler/residential_quota.py

@@ -63,7 +63,7 @@ def _run_sampling_docker(self):
             self.parent().docker_image,
             [
                 "ruby",
-                "resources/run_sampling.rb",
+                "samplers/quota/run_sampling.rb",
                 "-p",
                 self.cfg["project_directory"],
                 "-n",
@@ -100,7 +100,7 @@ def _run_sampling_apptainer(self):
             "{}:/outbind".format(os.path.dirname(self.csv_path)),
             self.parent().apptainer_image,
             "ruby",
-            "resources/run_sampling.rb",
+            "samplers/quota/run_sampling.rb",
             "-p",
             self.cfg["project_directory"],
             "-n",
@@ -113,11 +113,11 @@ def _run_sampling_apptainer(self):
         logger.debug("Apptainer sampling completed.")
         return self.csv_path
 
-    def _run_sampling_local_openstudio(self):
+    def _run_sampling_local(self):
         subprocess.run(
             [
-                self.parent().openstudio_exe(),
-                str(pathlib.Path("resources", "run_sampling.rb")),
+                "ruby",
+                str(pathlib.Path("samplers", "quota", "run_sampling.rb")),
                 "-p",
                 self.cfg["project_directory"],
                 "-n",

buildstockbatch/sampler/residential_stratified.py

@@ -0,0 +1,181 @@
+"""
+buildstockbatch.sampler.residential_stratified
+~~~~~~~~~~~~~~~
+This object contains the code required for generating the set of simulations to execute
+
+:author: Noel Merket, Ry Horsey
+:copyright: (c) 2020 by The Alliance for Sustainable Energy
+:license: BSD-3
+"""
+
+import docker
+import logging
+import os
+import pathlib
+import shutil
+import subprocess
+import sys
+import time
+import yaml
+
+from .base import BuildStockSampler
+from buildstockbatch.exc import ValidationError
+
+logger = logging.getLogger(__name__)
+
+
+class ResidentialStratifiedSampler(BuildStockSampler):
+    def __init__(
+        self,
+        parent,
+        n_datapoints,
+        segment_vars=[
+            "Federal Poverty Level",
+            "Geometry Floor Area Bin",
+            "Geometry Building Type RECS",
+            "Vintage",
+            "Heating Fuel",
+            "Sampling Region",
+        ],
+        segment_selection_sample_size=10000000,
+        num_samples_per_segment=8,
+    ):
+        """Residential Stratified Sampler
+
+        :param parent: BuildStockBatchBase object
+        :type parent: BuildStockBatchBase (or subclass)
+        :param n_datapoints: number of datapoints to sample
+        :type n_datapoints: int
+        """
+        super().__init__(parent)
+        self.validate_args(self.parent().project_filename, n_datapoints=n_datapoints)
+        self.n_datapoints = n_datapoints
+        self.sampler_config = self.create_sampler_config(
+            os.path.dirname(self.parent().project_filename),
+            segment_vars,
+            segment_selection_sample_size,
+            num_samples_per_segment,
+        )
+
+    @classmethod
+    def validate_args(cls, project_filename, **kw):
+        expected_args = set(["n_datapoints"])
+        for k, v in kw.items():
+            expected_args.discard(k)
+            if k == "n_datapoints":
+                if not isinstance(v, int):
+                    raise ValidationError("n_datapoints needs to be an integer")
+                if v <= 0:
+                    raise ValidationError("n_datapoints need to be >= 1")
+            elif k == "segment_vars":
+                pass
+            elif k == "segment_selection_sample_size":
+                pass
+            elif k == "num_samples_per_segment":
+                pass
+            else:
+                raise ValidationError(f"Unknown argument for sampler: {k}")
+        if len(expected_args) > 0:
+            raise ValidationError("The following sampler arguments are required: " + ", ".join(expected_args))
+        return True
+
+    @classmethod
+    def create_sampler_config(self, folderpath, segment_vars, segment_selection_sample_size, num_samples_per_segment):
+        data = {}
+        data["segment_vars"] = segment_vars
+        data["segment_selection_sample_size"] = segment_selection_sample_size
+        data["num_samples_per_segment"] = num_samples_per_segment
+        filename = pathlib.Path(folderpath) / "sampler_config.yaml"
+        with open(filename, "w") as file:
+            yaml.dump(data, file)
+        return filename
+
+    def _run_sampling_docker(self):
+        docker_client = docker.DockerClient.from_env()
+        tick = time.time()
+        extra_kws = {}
+        if sys.platform.startswith("linux"):
+            extra_kws["user"] = f"{os.getuid()}:{os.getgid()}"
+        container_output = docker_client.containers.run(
+            self.parent().docker_image,
+            [
+                "python",
+                "samplers/stratified/sampler/run_sampler.py",
+                "-p",
+                self.cfg["project_directory"],
+                "-n",
+                str(self.n_datapoints),
+                "-o",
+                "buildstock.csv",
+            ],
+            remove=True,
+            volumes={self.buildstock_dir: {"bind": "/var/simdata/openstudio", "mode": "rw"}},
+            name="buildstock_sampling",
+            **extra_kws,
+        )
+        tick = time.time() - tick
+        for line in container_output.decode("utf-8").split("\n"):
+            logger.debug(line)
+        logger.debug("Sampling took {:.1f} seconds".format(tick))
+        destination_filename = self.csv_path
+        if os.path.exists(destination_filename):
+            os.remove(destination_filename)
+        shutil.move(
+            os.path.join(self.buildstock_dir, "resources", "buildstock.csv"),
+            destination_filename,
+        )
+        return destination_filename
+
+    def _run_sampling_apptainer(self):
+        args = [
+            "apptainer",
+            "exec",
+            "--contain",
+            "--home",
+            "{}:/buildstock".format(self.buildstock_dir),
+            "--bind",
+            "{}:/outbind".format(os.path.dirname(self.csv_path)),
+            self.parent().apptainer_image,
+            "python",
+            "samplers/stratified/sampler/run_sampler.py",
+            "-p",
+            self.cfg["project_directory"],
+            "-n",
+            str(self.n_datapoints),
+            "-o",
+            "../../outbind/{}".format(os.path.basename(self.csv_path)),
+        ]
+        logger.debug(f"Starting apptainer sampling with command: {' '.join(args)}")
+        subprocess.run(args, check=True, env=os.environ, cwd=self.parent().output_dir)
+        logger.debug("Apptainer sampling completed.")
+        return self.csv_path
+
+    def _run_sampling_local(self):
+        subprocess.run(
+            [
+                "python",
+                str(pathlib.Path("samplers", "stratified", "sampler", "run_sampler.py")),
+                "sample",
+                "-p",
+                self.cfg["project_directory"],
+                "-n",
+                str(self.n_datapoints),
+                "-c",
+                self.sampler_config,
+                "-o",
+                "buildstock.csv",
+            ],
+            cwd=self.buildstock_dir,
+            check=True,
+        )
+        destination_filename = pathlib.Path(self.csv_path)
+        if destination_filename.exists():
+            os.remove(destination_filename)
+        shutil.move(
+            pathlib.Path(self.buildstock_dir, "resources", "buildstock.csv"),
+            destination_filename,
+        )
+        config_filename = pathlib.Path(self.sampler_config)
+        if config_filename.exists():
+            os.remove(config_filename)
+        return destination_filename