Merge branch 'main' of https://github.com/JessicaOjo/lm-evaluation-harness into africamgsm

lm_eval/tasks/afrimgsm/translate_direct/translate_direct_yaml

+# This file will be included in the generated language-specific task configs.
+# It doesn't have a yaml file extension as it is not meant to be imported directly
+# by the harness.
+group: translate_afrimgsm_direct
+dataset_path: masakhane/afrimgsm-translate-test
+dataset_name: null  # Overridden by language-specific config.
+output_type: generate_until
+test_split: test
+generation_kwargs:
+  until:
+    - "\n\n"
+    - "\n"
+  do_sample: false
+  temperature: 0.0
+target_delimiter: " "
+metric_list:
+  - metric: exact_match
+    aggregation: mean
+    higher_is_better: true
+    ignore_case: true
+    ignore_punctuation: true
+filter_list:
+  - name: "get-answer"
+    filter:
+      - function: "regex"
+        regex_pattern: "The answer is (\\-?[0-9\\.\\,]+)"
+      - function: "take_first"
+metadata:
+  version: 3.0

lm_eval/tasks/afrimgsm/utils.py

@@ -69,8 +69,8 @@ def main() -> None:
     )
     parser.add_argument(
         "--mode",
-        default="direct",
-        choices=["direct", "native-cot", "en-cot"],
+        default="native-cot",
+        choices=["direct","direct-native", "native-cot", "en-cot","translate-direct"],
         help="Mode of chain-of-thought",
     )
     args = parser.parse_args()