fix: fix multiapp tasks (#229)

Co-authored-by: adlsdztony <zzl0712@connect.hku.hk>
2025-07-03 21:53:58 +08:00
parent 7b2120c843
commit ac24ccce99
12 changed files with 57 additions and 28 deletions
--- a/desktop_env/evaluators/metrics/others.py
+++ b/desktop_env/evaluators/metrics/others.py
@@ -63,11 +63,13 @@ def compare_epub(result: str, expected: str) -> float:
    result_files: List[str] = process_epub(result)
    expected_files: List[str] = process_epub(expected)

-    metric: float = 1.
+    metric: float = 0.
    for f1, f2 in zip(result_files, expected_files):
        current_metric: float = diff_text_file(f1, f2)
        logger.debug("%s vs %s: %f", f1, f2, current_metric)
-        metric *= current_metric
+        metric += current_metric
+    if len(result_files) > 0:
+        metric /= len(result_files)
    return metric