Merge branch 'upstream' into 'mmlu-pro'

add tokenizer logs info (#1731)

See merge request shijie.yu/lm-evaluation-harness!4

lm_eval/tasks/afrimmlu/translate/afrimmlu_common_translate_yaml

+group:
+  - mmlu
+  - afrimmlu_translate
+task: null
+dataset_path: masakhane/afrimmlu-translate-test
+dataset_name: null
+output_type: multiple_choice
+test_split: test
+doc_to_text: !function utils.doc_to_text
+doc_to_target: "{{['A', 'B', 'C', 'D'].index(answer)}}"
+doc_to_choice: !function utils.doc_to_choice
+should_decontaminate: true
+doc_to_decontamination_query: "Question: {{question}}\nAnswer:"
+metric_list:
+  - metric: f1
+    aggregation: !function utils.weighted_f1_score
+    # aggregation: mean
+    average: weighted
+    hf_evaluate: true
+    higher_is_better: True
+    ignore_case: true
+    ignore_punctuation: true
+    regexes_to_ignore:
+      - ","
+      - "\\$"
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true
+    ignore_case: true
+    ignore_punctuation: true
+    regexes_to_ignore:
+      - ","
+      - "\\$"
+metadata:
+  version: 1.0

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_amh.yaml

+dataset_name: amh
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_amh

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_eng.yaml

+dataset_name: eng
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_eng

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_ewe.yaml

+dataset_name: ewe
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_ewe

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_fra.yaml

+dataset_name: fra
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_fra

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_hau.yaml

+dataset_name: hau
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_hau

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_ibo.yaml

+dataset_name: ibo
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_ibo

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_kin.yaml

+dataset_name: kin
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_kin

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_lin.yaml

+dataset_name: lin
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_lin

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_lug.yaml

+dataset_name: lug
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_lug

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_orm.yaml

+dataset_name: orm
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_orm

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_sna.yaml

+dataset_name: sna
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_sna

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_sot.yaml

+dataset_name: sot
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_sot

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_swa.yaml

+dataset_name: swa
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_swa

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_twi.yaml

+dataset_name: twi
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_twi

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_wol.yaml

+dataset_name: wol
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_wol

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_xho.yaml

+dataset_name: xho
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_xho

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_yor.yaml

+dataset_name: yor
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_yor

lm_eval/tasks/afrimmlu/translate/afrimmlu_translate_zul.yaml

+dataset_name: zul
+include: afrimmlu_common_translate_yaml
+task: afrimmlu_translate_zul

lm_eval/tasks/afrimmlu/translate/utils.py

+from sklearn.metrics import f1_score
+
+
+def doc_to_choice(doc):
+    choices = eval(doc["choices"])
+    return choices
+
+
+def doc_to_text(doc):
+    output = """You are a highly knowledgeable and intelligent artificial intelligence
+                model answers multiple-choice questions about '{subject}'
+
+                Question: '''{question}'''
+
+                Choices:
+                        A: ''{choice1}'''
+                        B: ''{choice2}'''
+                        C: ''{choice3}'''
+                        D: ''{choice4}'''
+
+                Answer:  """
+
+    choices = eval(doc["choices"])
+    text = output.format(
+        subject=doc["subject"],
+        question=doc["question"],
+        choice1=choices[0],
+        choice2=choices[1],
+        choice3=choices[2],
+        choice4=choices[3],
+    )
+    return text
+
+
+def weighted_f1_score(items):
+    unzipped_list = list(zip(*items))
+    golds = unzipped_list[0]
+    preds = unzipped_list[1]
+    fscore = f1_score(golds, preds, average="weighted")
+    return fscore