lzy/sci-gui-agent-benchmark
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

ver Mar8th

Browse Source 
fixed a task and a metric
This commit is contained in:
David Chang
2024-03-08 13:28:34 +08:00
parent ddcafac30f
commit ce23f3dab4
3 changed files with 33 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
desktop_env/evaluators/metrics/table.py
View File

@@ -194,7 +194,7 @@ def compare_table(result: str, expected: str = None, **options) -> float:
# sheet_idx1: as sheet_idx0
# rules: list of dict, each dict is like
# { "range": ["A1:B6", "C2:E5"],
# "type": "includes" | "includes_by" | "fuzzy_match" | "exact_match", # 0 includes 1, 0 includes_by 1
# "type": "includes" | "included_by" | "fuzzy_match" | "exact_match", # 0 includes 1, 0 includes_by 1
# "threshold": 85, // for fuzzy match
# "ignore_case": true | false,
# "ignore_chars": " ()", # filtered out
@@ -232,9 +232,9 @@ def compare_table(result: str, expected: str = None, **options) -> float:
value2 = value2.lower()
if rl["type"]=="includes":
metric: bool = value1 in value2
elif rl["type"]=="includes_by":
metric: bool = value2 in value1
elif rl["type"]=="included_by":
metric: bool = value1 in value2
elif rl["type"]=="fuzzy_match":
metric: bool = fuzz.ratio(value1, value2) >= rl.get("threshold", 85.)
elif rl["type"]=="exact_match":