Adding ast library

coord_eval.py

@@ -16,6 +16,7 @@
 import evaluate
 import datasets
 import re
+import ast
 
 
 # TODO: Add BibTeX citation
@@ -45,7 +46,7 @@ Returns:
     accuracy: 1 if coordinates predicted are d distant from gold ones, O otherwise.
 Examples:
     >>> my_new_module = evaluate.load("rfr2003/coord_eval")
-    >>> results = my_new_module.compute(generations=["(12.7, 67.8)", "(16.7, 89.6)"], golds=[[12.7, 67.8], [10.9, 80.6]], d=20)
+    >>> results = my_new_module.compute(generations=["(12.7, 67.8)", "(16.7, 89.6)"], golds=[[12.7, 67.8], [10.9, 80.6]], d_range=20)
     >>> print(results)
     {'coord_accuracy': 0.5}
 """
@@ -119,7 +120,7 @@ class Coord_eval(evaluate.Metric):
                 if matches:
                     match = matches[0]
                     f_ans = (float(match[0].replace(',', '.').replace(' ', '')), float(match[3].replace(',', '.').replace(' ', '')))
-                    correct += accuracy_coord(f_ans, f_gold, d_range)
+                    correct += self._accuracy_coord(f_ans, f_gold, d_range)
                             
             total += 1
 

tests.py

@@ -2,13 +2,13 @@ test_cases = [
     {
         "generations": ["(12.7, 67.8)", "(16.7, 89.6)"],
         "golds": [[12.7, 67.8], [10.9, 80.6]],
-        "d": 20,
+        "d_range": 20,
         "result": {"coord_accuracy": 0.5}
     },
     {
         "generations": ["(12.7, 67.8)", "(16.7, 89.6)"],
         "golds": [[12.7, 67.8], [10.9, 80.6]],
-        "d": 1000000,
+        "d_range": 1000000,
         "result": {"coord_accuracy": 1}
     },
 ]