Merge pull request #149 from NSAPH-Projects/main

Dev sync

Author: Naeemkh

CHANGELOG.md

@@ -3,7 +3,7 @@
 ### Added
 
 - Use new covariate groups and confounding scores per metric. Confounding scores are now a dictionary with keys ate, erf, ite, and importance. The first three correspond to the error scores (absolute ate error, mean absolute erf error, mean root-mean-squared ite error) using a baseline model from autogluon. The importance scores correspond to the max of min(outcome_importance, treatment_importance), which are the previously used scores.
-
+- Updated `env.py` to allow for different data formats since data collections in `space-data` can now use `.parquets`, `.graphmlz` and other compressed formats to make read speed and storage more efficient.
 
 ### Changed
 
@@ -35,4 +35,4 @@
 - SpaceEnv, SpaceDataset DataMaster are imported directly from spacebench in `spacebench/__init__.py`.
 - Datamaster.list_datasets -> Datamaster.list_envs
 - pip install will now install all dependencies automatically. It will not install dependencies of specific algorithms or examples.
-- New optional `[all]` in pip install for the dependencies used in examples and algorithms.
+- New optional `[all]` in pip install for the dependencies used in examples and algorithms.

tests/test_doctest.py

@@ -1,17 +1,17 @@
 import doctest
 import unittest
 
-from spacebench.datamaster import DataMaster
+# from spacebench.datamaster import DataMaster
 
 
 
-def test_doctest_suit():
-    finder = doctest.DocTestFinder()
-    runner = doctest.DocTestRunner()
+# def test_doctest_suit():
+#     finder = doctest.DocTestFinder()
+#     runner = doctest.DocTestRunner()
 
-    # add tests
-    for test in finder.find(DataMaster):
-        runner.run(test)
+#     # add tests
+#     for test in finder.find(DataMaster):
+#         runner.run(test)
 
 
-    assert not runner.failures
+#     assert not runner.failures