Merge branch 'main' into feature/eval_from_config

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_orm.yaml

+# Generated by utils.py
+dataset_name: orm
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_orm_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_sna.yaml

+# Generated by utils.py
+dataset_name: sna
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_sna_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_sot.yaml

+# Generated by utils.py
+dataset_name: sot
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_sot_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_swa.yaml

+# Generated by utils.py
+dataset_name: swa
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_swa_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_twi.yaml

+# Generated by utils.py
+dataset_name: twi
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_twi_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_wol.yaml

+# Generated by utils.py
+dataset_name: wol
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_wol_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_xho.yaml

+# Generated by utils.py
+dataset_name: xho
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_xho_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_yaml

+tag:
+    - afrixnli_tasks
+    - afrixnli_tasks_prompt_5
+dataset_path: masakhane/afrixnli
+dataset_name: null
+output_type: multiple_choice
+validation_split: validation
+test_split: test
+fewshot_split: validation
+doc_to_target: !function utils.doc_to_target
+doc_to_choice:
+  - "true"
+  - "inconclusive"
+  - "false"
+should_decontaminate: true
+doc_to_decontamination_query: premise
+metric_list:
+  - metric: f1
+    aggregation: !function utils.weighted_f1_score
+    average: weighted
+    higher_is_better: True
+    ignore_case: true
+    ignore_punctuation: true
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true
+    ignore_case: true
+    ignore_punctuation: true
+metadata:
+  version: 1.0

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_yor.yaml

+# Generated by utils.py
+dataset_name: yor
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_yor_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/afrixnli_zul.yaml

+# Generated by utils.py
+dataset_name: zul
+doc_to_text: "Based on the given statement, is the following claim 'true', 'false',\
+  \ or 'inconclusive'. \nStatement: {{premise}} \nClaim: {{hypothesis}}"
+include: afrixnli_yaml
+task: afrixnli_zul_prompt_5

lm_eval/tasks/afrixnli/direct/prompt_5/utils.py

+from lm_eval.utils import weighted_f1_score
+
+
+def doc_to_target(doc):
+    replacements = {0: "true", 1: "false", 2: "inconclusive"}
+    return replacements[doc["label"]]

lm_eval/tasks/afrixnli/gen_utils.py

+import argparse
+import os
+
+import yaml
+
+
+class FunctionTag:
+    def __init__(self, value):
+        self.value = value
+
+
+def prompt_func(mode, lang):
+    prompt_map = {
+        "prompt_1": "Please identify whether the premise entails or contradicts the hypothesis in the following premise "
+        "and hypothesis. The answer should be exact entailment, contradiction, or neutral.\n\nPremise: {premise}\nHypothesis: {hypothesis}\n\n"
+        "Is it entailment, contradiction, or neutral?",
+        "prompt_3": f"Given the following premise and hypothesis in {lang}, identify if the premise entails, contradicts, "
+        f"or is neutral towards the hypothesis. Please respond with exact 'entailment', 'contradiction', or 'neutral'. \n\n"
+        "Premise: {{premise}} \nHypothesis: {{hypothesis}}",
+        "prompt_4": f"You are an expert in Natural Language Inference (NLI) specializing in the {lang} language.\n"
+        f"Analyze the premise and hypothesis given in {lang}, and determine the relationship between them.\n "
+        f"Respond with one of the following options: 'entailment', 'contradiction', or 'neutral'. \n\n"
+        "Premise: {{premise}} \nHypothesis: {{hypothesis}}",
+        "prompt_5": "Based on the given statement, is the following claim 'true', 'false', or 'inconclusive'. \n"
+        "Statement: {{premise}} \nClaim: {{hypothesis}}",
+    }
+    return prompt_map[mode]
+
+
+def gen_lang_yamls(output_dir: str, overwrite: bool, mode: str) -> None:
+    """
+    Generate a yaml file for each language.
+
+    :param output_dir: The directory to output the files to.
+    :param overwrite: Whether to overwrite files if they already exist.
+    """
+    err = []
+    languages = {
+        "eng": "English",
+        "amh": "Amharic",
+        "ibo": "Igbo",
+        "fra": "French",
+        "sna": "chiShona",
+        "wol": "Wolof",
+        "ewe": "Ewe",
+        "lin": "Lingala",
+        "lug": "Luganda",
+        "xho": "isiXhosa",
+        "kin": "Kinyarwanda",
+        "twi": "Twi",
+        "zul": "Zulu",
+        "orm": "Oromo",
+        "yor": "Yoruba",
+        "hau": "Hausa",
+        "sot": "Sesotho",
+        "swa": "Swahili",
+    }
+
+    for lang in languages.keys():
+        try:
+            file_name = f"afrixnli_{lang}.yaml"
+            task_name = f"afrixnli_{lang}_{mode}"
+            yaml_template = "afrixnli_yaml"
+            if output_dir.split("/")[-1] == "translate":
+                file_name = f"afrixnli_translate_{lang}.yaml"
+                task_name = f"afrixnli_translate_{lang}_{mode}"
+                yaml_template = "afrixnli_translate_yaml"
+            if int(mode.split("_")[-1]) == 1 or int(mode.split("_")[-1]) > 2:
+                yaml_details = {
+                    "include": yaml_template,
+                    "task": task_name,
+                    "dataset_name": lang,
+                    "doc_to_text": prompt_func(mode, languages[lang]),
+                }
+            else:
+                yaml_details = {
+                    "include": yaml_template,
+                    "task": task_name,
+                    "dataset_name": lang,
+                }
+            os.makedirs(f"{output_dir}/{mode}", exist_ok=True)
+            with open(
+                f"{output_dir}/{mode}/{file_name}",
+                "w" if overwrite else "x",
+                encoding="utf8",
+            ) as f:
+                f.write("# Generated by utils.py\n")
+                yaml.dump(
+                    yaml_details,
+                    f,
+                    allow_unicode=True,
+                )
+        except FileExistsError:
+            err.append(file_name)
+
+    if len(err) > 0:
+        raise FileExistsError(
+            "Files were not created because they already exist (use --overwrite flag):"
+            f" {', '.join(err)}"
+        )
+
+
+def main() -> None:
+    """Parse CLI args and generate language-specific yaml files."""
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--overwrite",
+        default=True,
+        action="store_true",
+        help="Overwrite files if they already exist",
+    )
+    parser.add_argument(
+        "--output-dir",
+        default="./translate",
+        help="Directory to write yaml files to",
+    )
+    parser.add_argument(
+        "--mode",
+        default="prompt_5",
+        choices=["prompt_1", "prompt_2", "prompt_3", "prompt_4", "prompt_5"],
+        help="Prompt number",
+    )
+    args = parser.parse_args()
+
+    gen_lang_yamls(output_dir=args.output_dir, overwrite=args.overwrite, mode=args.mode)
+
+
+if __name__ == "__main__":
+    main()

lm_eval/tasks/afrixnli/translate/afrixnli_tt.yaml

+group: afrixnli_tt-irokobench
+task:
+  - afrixnli_tt_tasks
+aggregate_metric_list:
+  - metric: acc
+    aggregation: mean
+    weight_by_size: true
+metadata:
+  version: 2

lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_amh.yaml

+# Generated by utils.py
+dataset_name: amh
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_amh_prompt_1

lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_ewe.yaml

+# Generated by utils.py
+dataset_name: ewe
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ewe_prompt_1

lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_fra.yaml

+# Generated by utils.py
+dataset_name: fra
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_fra_prompt_1

lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_hau.yaml

+# Generated by utils.py
+dataset_name: hau
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_hau_prompt_1

lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_ibo.yaml

+# Generated by utils.py
+dataset_name: ibo
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_ibo_prompt_1

lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_kin.yaml

+# Generated by utils.py
+dataset_name: kin
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_kin_prompt_1

lm_eval/tasks/afrixnli/translate/prompt_1/afrixnli_translate_lin.yaml

+# Generated by utils.py
+dataset_name: lin
+doc_to_text: 'Please identify whether the premise entails or contradicts the hypothesis
+  in the following premise and hypothesis. The answer should be exact entailment,
+  contradiction, or neutral.
+
+
+  Premise: {premise}
+
+  Hypothesis: {hypothesis}
+
+
+  Is it entailment, contradiction, or neutral?'
+include: afrixnli_translate_yaml
+task: afrixnli_translate_lin_prompt_1