1) added a script that can be called from command line to run popgen; accordingly modified .gitignore file, and the setup,py file, 2) cleaned configuration elements, 3) altered the way input files are specified, now, input files are to be specified in the project folder, 4) deleted any references to configuration elements that are not used, 5) commented out a print statement in the reweighting procedure

Author: karthikcharan83

.gitignore

@@ -10,7 +10,6 @@ dist
 build
 eggs
 parts
-bin
 var
 sdist
 develop-eggs
@@ -39,4 +38,4 @@ output/*.html
 output/*/index.html
 
 # Sphinx
-docs/_build
+docs/_build

bin/popgen_run

@@ -0,0 +1,40 @@
+#!/usr/bin/env python
+
+import sys
+import time
+
+from popgen import Project
+
+
+class ArgumentsError(Exception):
+    pass
+
+
+def run():
+    """
+    Runs the PopGen program to generate a synthetic population.
+    Please refer to PopGen documentation on the GitHub site
+    for guidance on setting up the configuration file.
+    """
+    args = sys.argv[1:]
+
+    print args
+
+    if len(args) != 1:
+        raise ArgumentsError(
+            """The module accepts only one argument which is the location of
+            the configuration file.
+
+            Example usage on linux or Mac:
+                popgen_run /location/of/file.yaml
+
+            Example usage on Windows:
+                popgen_run c:/location/of/file.yaml""")
+
+    t = time.time()
+    p_obj = Project(config_loc=args[0])
+    p_obj.load_project()
+    p_obj.run_scenarios()
+    print "Time it took: %.4f" % (time.time() - t)
+
+run()

popgen/data.py

@@ -30,6 +30,7 @@ def __init__(self, config):
         self.region_ids = None
         self.sample_geo_ids = None
         self._inputs_config = self.config.project.inputs
+        self.location = os.path.abspath(self.config.project.location)
 
     def load_data(self):
 
@@ -54,7 +55,8 @@ def get_data(self, config, header=0):
         # print config_dict, type(config_dict)
         data_dict = {}
         for item in config_dict:
-            full_location = os.path.abspath(config_dict[item])
+            filename = config_dict[item]
+            full_location = os.path.join(self.location, filename)
             data_dict[item] = pd.read_csv(full_location, index_col=0,
                                           header=header)
             data_dict[item].loc[:,

popgen/project.py

@@ -38,7 +38,6 @@ def _load_config(self):
 
     def _populate_project_properties(self):
         self.name = self._config.project.name
-        self.synthesize = self._config.project.synthesize
         self.location = os.path.abspath(self._config.project.location)
 
     def _load_data(self):

popgen/reweighting.py

@@ -157,8 +157,8 @@ def run_reweighting(self, region_constraints, geo_constraints):
                             geo_constraints.loc[geo_id])
                         pass
                     # raw_input("Geo done %s" %geo_id)
-                # raw_input ("One outer iteration complete in %.4f" %
-                #            (time.time() - t))
+                # print ("\t\t\tOne outer iteration complete in %.4f" %
+                #       (time.time() - t))
             self._populate_sample_weights(sample_weights, region_id, geo_ids)
             # print self.average_deviations
             print "\tsample_weights sum:", sample_weights.sum()
@@ -262,6 +262,7 @@ def _calculate_populate_average_deviation(
             self, geo_id, iter, sample_weights, constraints):
         diff_sum = 0
 
+        sample_weights = np.array(sample_weights, order="C")
         for column in constraints.index:
             weighted_sum = sample_weights.dot(self.geo_contrib[column])
             diff_sum += (np.abs(weighted_sum - constraints[column]) /

setup.py

@@ -38,6 +38,7 @@
     package_data={'popgen': [
         '../tutorials/1_basic_popgen_setup/*.csv',
         '../tutorials/1_basic_popgen_setup/*.yaml']},
+    scripts=['bin/popgen_run'],
     include_package_data=True,
     setup_requires=['numpy==1.9.2'],
     install_requires=[

tutorials/1_basic_popgen_setup/configuration.yaml

@@ -1,7 +1,6 @@
 project:
- synthesize: True
  name: example
- location: ../tutorials/1_basic_popgen_setup/
+ location: ./tutorials/1_basic_popgen_setup/
  inputs:
   entities: [household, person]
   housing_entities: [household]
@@ -14,23 +13,21 @@ project:
    sample_geo: sample_geo
   location:
    geo_corr_mapping:
-    geo_to_sample: ../tutorials/1_basic_popgen_setup/geo_sample_mapping.csv
-    region_to_sample: ../tutorials/1_basic_popgen_setup/region_sample_mapping.csv
-    region_to_geo: ../tutorials/1_basic_popgen_setup/region_geo_mapping.csv
+    geo_to_sample: geo_sample_mapping.csv
+    region_to_sample: region_sample_mapping.csv
+    region_to_geo: region_geo_mapping.csv
    sample:
-    household: ../tutorials/1_basic_popgen_setup/household_sample.csv
-    person: ../tutorials/1_basic_popgen_setup/person_sample.csv
+    household: household_sample.csv
+    person: person_sample.csv
    marginals:
-    geo:
-     household: ../tutorials/1_basic_popgen_setup/household_marginals.csv
-     person: ../tutorials/1_basic_popgen_setup/person_marginals.csv
     region:
-     household: ../tutorials/1_basic_popgen_setup/region_household_marginals.csv
-     person: ../tutorials/1_basic_popgen_setup/region_person_marginals.csv
-
+     household: region_household_marginals.csv
+     person: region_person_marginals.csv
+    geo:
+     household: household_marginals.csv
+     person: person_marginals.csv
  scenario:
-  - description: all_controls_entropy
-    apply_region_controls: True
+  - description: all_controls_ipu
     control_variables:
      region:
       household: [rhhldtype]
@@ -46,7 +43,7 @@ project:
       rounding_procedure: bucket
       archive_performance_frequency: 1
      reweighting:
-      procedure: entropy
+      procedure: ipu
       tolerance: 0.0001
       inner_iterations: 1
       outer_iterations: 1000

tutorials/1_basic_popgen_setup/configuration_multiple_scenarios.yaml

@@ -1,5 +1,4 @@
 project:
- synthesize: True
  name: example
  location: ../demo/example/
  inputs:
@@ -27,7 +26,6 @@ project:
     region:
      household: ../demo/example/region_household_marginals.csv
      person: ../demo/example/region_person_marginals.csv
-
  scenario:
   - description: derived_2_entropy
     apply_region_controls: True