Change COMA API to have "use_instances" as keyword instead of "strate…

…gy" (#60) * Change COMA API to have "use_instances" as keyword * Use keyword anyway for example and readme, to show users the option exists --------- Co-authored-by: Shaad Alaka <[email protected]>
delftdata · Oct 10, 2023 · 3b67275 · 3b67275
1 parent 06fdc82
commit 3b67275
Show file tree

Hide file tree

Showing 5 changed files with 8 additions and 8 deletions.
diff --git a/README.md b/README.md
@@ -93,7 +93,7 @@ df1 = pd.read_csv(d1_path)
 df2 = pd.read_csv(d2_path)
 
 # Instantiate matcher and run
-matcher = Coma(strategy="COMA_OPT")
+matcher = Coma(use_instances=True)
 matches = valentine_match(df1, df2, matcher)
 
 print(matches)

diff --git a/examples/valentine_example.py b/examples/valentine_example.py
@@ -15,7 +15,7 @@ def main():
     # Instantiate matcher and run
     # Coma requires java to be installed on your machine
     # If java is not an option, all the other algorithms are in Python (e.g., Cupid)
-    matcher = Coma(strategy="COMA_OPT")
+    matcher = Coma(use_instances=False)
     matches = valentine_match(df1, df2, matcher)
 
     # If ground truth available valentine could calculate the metrics

diff --git a/tests/test_algorithms.py b/tests/test_algorithms.py
@@ -12,11 +12,11 @@ class TestAlgorithms(unittest.TestCase):
 
     def test_coma(self):
         # Test the schema variant of coma
-        coma_matcher_schema = Coma(strategy="COMA_OPT")
+        coma_matcher_schema = Coma(use_instances=False)
         matches_coma_matcher_schema = coma_matcher_schema.get_matches(d1, d2)
         assert len(matches_coma_matcher_schema) > 0  # Check that it actually produced output
         # Test the instance variant of coma
-        coma_matcher_instances = Coma(strategy="COMA_OPT_INST")
+        coma_matcher_instances = Coma(use_instances=True)
         matches_coma_matcher_instances = coma_matcher_instances.get_matches(d1, d2)
         assert len(matches_coma_matcher_instances) > 0  # Check that it actually produced output
         # Assume the Schema and instance should provide different results

diff --git a/tests/test_valentine.py b/tests/test_valentine.py
@@ -12,7 +12,7 @@ class TestValentine(unittest.TestCase):
     def test_match(self):
         assert not DataframeTable(df1, name='df1_name').is_empty
         assert not DataframeTable(df2, name='df2_name').is_empty
-        matches = valentine_match(df1, df2, Coma(strategy="COMA_OPT_INST"))
+        matches = valentine_match(df1, df2, Coma(use_instances=True))
         assert len(matches) > 0
         try:
             valentine_match(df1, df2, None)
@@ -22,7 +22,7 @@ def test_match(self):
             assert False
 
     def test_metrics(self):
-        matches = valentine_match(df1, df2, Coma(strategy="COMA_OPT_INST"))
+        matches = valentine_match(df1, df2, Coma(use_instances=True))
         golden_standard = [('Cited by', 'Cited by'),
                            ('Authors', 'Authors'),
                            ('EID', 'EID')]

diff --git a/valentine/algorithms/coma/coma.py b/valentine/algorithms/coma/coma.py
@@ -14,10 +14,10 @@ class Coma(BaseMatcher):
 
     def __init__(self,
                  max_n: int = 0,
-                 strategy: str = "COMA_OPT",
+                 use_instances: bool = False,
                  java_xmx: str = "1024m"):
         self.__max_n = int(max_n)
-        self.__strategy = strategy
+        self.__strategy = "COMA_OPT_INST" if use_instances else "COMA_OPT"
         self.__java_XmX = java_xmx
 
     def get_matches(self,