Files
sci-gui-agent-benchmark/desktop_env/evaluators/metrics/vscode.py

54 lines
1.2 KiB
Python

def compare_text_file(actual: str, expected: str, **options) -> float:
"""
Args:
actual (str): path to result text file
expected (str): path to gold text file
Return:
float: the score
"""
with open(actual) as f1:
actual_text = f1.read()
with open(expected) as f2:
expected_text = f2.read()
if actual_text == expected_text:
return 1.0
return 0.0
def compare_config(actual: str, expected: str, **options) -> float:
"""
Args:
actual (str): path to result text file
expected (str): gold string
Return:
float: the score
"""
with open(actual) as f1:
actual_text = f1.read()
if actual_text == expected:
return 1.0
return 0.0
def compare_answer(actual: str, expected: str, **options) -> float:
"""
Args:
actual (str): result string
expected (str): gold string
Return:
float: the score
"""
if actual == expected:
return 1.0
# TODO: can use text embedding to get non-zero return
return 0.0
if __name__ == '__main__':
print(compare_text_file("README.md", "README.md"))