We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 9044da9 commit a0cc11fCopy full SHA for a0cc11f
1 file changed
evalbench/scorers/trajectorymatcher.py
@@ -7,10 +7,11 @@
7
from typing import Tuple, Any
8
from scorers import comparator
9
10
+
11
class TrajectoryMatcher(comparator.Comparator):
12
"""
13
TrajectoryMatcher class implements the Comparator base class for checking tool execution trajectories.
-
14
15
It checks if the sequence of executed tools matches the expected trajectory.
16
17
@@ -35,11 +36,11 @@ def compare(
35
36
) -> Tuple[float, str]:
37
38
Compares expected trajectory (golden) with actual executed tools (generated).
39
40
Args:
41
golden_execution_result: List of expected tool names (strings).
42
generated_execution_result: List of actually executed tool names (strings).
43
44
Returns:
45
Tuple (score, explanation)
46
0 commit comments