remove trust-remote-code in configs; fix escape sequences (#3180)

* remove trust-remote-code

* add W605 rule

lm_eval/tasks/race/race.yaml

@@ -12,5 +12,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 2.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/agi_eval/non_greedy_robustness_agieval_aqua_rat.yaml

@@ -32,5 +32,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/agi_eval/option_order_robustness_agieval_aqua_rat.yaml

@@ -43,5 +43,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/agi_eval/prompt_robustness_agieval_aqua_rat.yaml

@@ -61,5 +61,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/math/math_grader.py

@@ -267,12 +267,12 @@ def normalize_answer_string(expr: str) -> str:
         "\\\\textit",
     ]:
         expr = expr.replace(surround_str, "")
-        pattern = f"^{surround_str}" + "\{(?P<text>.+?)\}$"
+        pattern = f"^{surround_str}" + r"\{(?P<text>.+?)\}$"
         m = re.search(pattern, expr)
         if m is not None:
             expr = m.group("text")
 
-    expr = expr.replace("\!", "")
+    expr = expr.replace(r"\!", "")
     expr = expr.replace("\\%", "%")
     expr = expr.replace("\\$", "$")
     expr = expr.replace("$", "")
@@ -305,7 +305,7 @@ def normalize_answer_string(expr: str) -> str:
         "p.m.",
         "PM",
     ]:
-        expr = re.sub(f"{unit}(es)?(s)? *(\^[0-9]+)?", "", expr)
+        expr = re.sub(rf"{unit}(es)?(s)? *(\^[0-9]+)?", "", expr)
 
     if "day" in expr:
         days = [
@@ -326,7 +326,7 @@ def normalize_answer_string(expr: str) -> str:
         if not weekday_expressed:
             expr = re.sub("day(s)?", "", expr)
 
-    expr = re.sub("\^ *\\\\circ", "", expr)
+    expr = re.sub("\\^ *\\\\circ", "", expr)
 
     if len(expr) > 0 and expr[0] == "{" and expr[-1] == "}":
         expr = expr[1:-1]

lm_eval/tasks/score/math/non_greedy_robustness_math_algebra.yaml

@@ -32,5 +32,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/math/prompt_robustness_math_algebra.yaml

@@ -62,5 +62,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/mmlu_pro/score_non_greedy_robustness_mmlu_pro.yaml

@@ -34,5 +34,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/mmlu_pro/score_option_order_robustness_mmlu_pro.yaml

@@ -63,5 +63,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/score/mmlu_pro/score_prompt_robustness_mmlu_pro.yaml

@@ -63,5 +63,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/spanish_bench/flores_es/_flores_common_yaml

@@ -23,5 +23,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/unscramble/anagrams1.yaml

@@ -18,5 +18,3 @@ metric_list:
     ignore_punctuation: false
 metadata:
   version: 2.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/unscramble/anagrams2.yaml

@@ -18,5 +18,3 @@ metric_list:
     ignore_punctuation: false
 metadata:
   version: 2.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/unscramble/cycle_letters.yaml

@@ -18,5 +18,3 @@ metric_list:
     ignore_punctuation: false
 metadata:
   version: 2.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/unscramble/random_insertion.yaml

@@ -18,5 +18,3 @@ metric_list:
     ignore_punctuation: false
 metadata:
   version: 2.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/wikitext/wikitext.yaml

@@ -16,5 +16,3 @@ metric_list:
   - metric: bits_per_byte
 metadata:
   version: 2.0
-dataset_kwargs:
-  trust_remote_code: true

lm_eval/tasks/winogrande/default.yaml

@@ -15,5 +15,3 @@ metric_list:
     higher_is_better: true
 metadata:
   version: 1.0
-dataset_kwargs:
-  trust_remote_code: true

pyproject.toml

@@ -106,7 +106,7 @@ plugins.md029.allow_extended_start_values = true # ol-prefix
 plugins.md034.enabled = false # no-bare-urls
 
 [tool.ruff.lint]
-extend-select = ["I"]
+extend-select = ["I", "W605"]
 
 [tool.ruff.lint.isort]
 lines-after-imports = 2