add chinese version of humaneval, mbpp (#743)

* add chinese_version of humaneval,mbpp

* add humaneval&mbpp gen.py

* minor fix

* minor add

---------

Co-authored-by: yingfhu <[email protected]>

configs/datasets/humaneval_cn/humaneval_cn_gen.py

@@ -0,0 +1,4 @@
+from mmengine.config import read_base
+
+with read_base():
+    from .humaneval_cn_gen_6313aa import humaneval_cn_datasets  # noqa: F401, F403

configs/datasets/humaneval_cn/humaneval_cn_gen_6313aa.py

@@ -0,0 +1,37 @@
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.datasets import HumanevalDataset, HumanEvaluator, humaneval_postprocess_v2
+
+humaneval_reader_cfg = dict(
+    input_columns=['prompt'], output_column='task_id', train_split='test')
+
+# TODO: allow empty output-column
+humaneval_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(round=[
+            dict(
+                role='HUMAN',
+                prompt='完成以下Python代码任务:\n{prompt}'),
+        ])),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer, max_out_len=512))
+
+humaneval_eval_cfg = dict(
+    evaluator=dict(type=HumanEvaluator),
+    pred_role='BOT',
+    k=[1, 10, 100],  # the parameter only for humaneval
+    pred_postprocessor=dict(type=humaneval_postprocess_v2),
+)
+
+humaneval_cn_datasets = [
+    dict(
+        abbr='openai_humaneval_cn',
+        type=HumanevalDataset,
+        path='./data/humaneval_cn/human-eval-cn-v2-20210705.jsonl',
+        reader_cfg=humaneval_reader_cfg,
+        infer_cfg=humaneval_infer_cfg,
+        eval_cfg=humaneval_eval_cfg)
+
+]

configs/datasets/humaneval_cn/humaneval_cn_passk_gen_6313aa.py

@@ -0,0 +1 @@
+./humaneval_cn_gen_6313aa.py

configs/datasets/humaneval_cn/humaneval_cn_repeat10_gen_6313aa.py

@@ -0,0 +1,38 @@
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.datasets import HumanevalDataset, HumanEvaluator, humaneval_postprocess_v2
+
+humaneval_reader_cfg = dict(
+    input_columns=['prompt'], output_column='task_id', train_split='test')
+
+# TODO: allow empty output-column
+humaneval_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(round=[
+            dict(
+                role='HUMAN',
+                prompt='完成以下Python代码任务:\n{prompt}'),
+        ])),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer, max_out_len=512))
+
+humaneval_eval_cfg = dict(
+    evaluator=dict(type=HumanEvaluator),
+    pred_role='BOT',
+    k=[1, 10, 100],  # the parameter only for humaneval
+    pred_postprocessor=dict(type=humaneval_postprocess_v2),
+)
+
+humaneval_cn_datasets = [
+    dict(
+        abbr='openai_humaneval_cn_pass10',
+        type=HumanevalDataset,
+        path='./data/humaneval_cn/human-eval-cn-v2-20210705.jsonl',
+        num_repeats=10,
+        reader_cfg=humaneval_reader_cfg,
+        infer_cfg=humaneval_infer_cfg,
+        eval_cfg=humaneval_eval_cfg)
+
+]

configs/datasets/mbpp/sanitized_mbpp_gen_1e1056.py

@@ -4,7 +4,7 @@
 from opencompass.datasets import SanitizedMBPPDataset, MBPPEvaluator
 
 sanitized_mbpp_reader_cfg = dict(
-    input_columns=['text', 'test_list'], output_column='test_list')
+    input_columns=['text', 'test_list'], output_column='test_list_2')
 
 sanitized_mbpp_infer_cfg = dict(
     prompt_template=dict(

configs/datasets/mbpp_cn/mbpp_cn_gen.py

@@ -0,0 +1,4 @@
+from mmengine.config import read_base
+
+with read_base():
+    from .mbpp_cn_gen_1d1481 import mbpp_cn_datasets  # noqa: F401, F403

configs/datasets/mbpp_cn/mbpp_cn_gen_1d1481.py

@@ -0,0 +1,64 @@
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.datasets import MBPPDataset, MBPPEvaluator
+
+mbpp_reader_cfg = dict(
+    input_columns=['text', 'test_list'], output_column='test_list_2')
+
+mbpp_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            round=[
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个函数，从给定的两个元组列表中查找相似的元素。 你的代码应该通过这些测试：\n\n assert similar_elements((3, 4, 5, 6),(5, 7, 4, 10)) == (4, 5)\n assert similar_elements((1, 2, 3, 4),(5, 4, 3, 7)) == (3, 4) \n assert similar_elements((11, 12, 14, 13),(17, 15, 14, 13)) == (13, 14) \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'def similar_elements(test_tup1, test_tup2):\r\n  res = tuple(set(test_tup1) & set(test_tup2))\r\n  return (res)' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个 Python 函数来识别一个整数是否不是素数。 你的代码应该通过这些测试：\n\n assert is_not_prime(2) == False \n assert is_not_prime(10) == True \n assert is_not_prime(35) == True \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'import math\r\ndef is_not_prime(n):\r\n    result = False\r\n    for i in range(2,int(math.sqrt(n)) + 1):\r\n        if n % i == 0:\r\n            result = True\r\n    return result' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个函数，使用堆队列算法从给定的数字列表中查找最大整数。 你的代码应该通过这些测试：\n\n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],3)==[85, 75, 65] \n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],2)==[85, 75] \n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],5)==[85, 75, 65, 58, 35] \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'import heapq as hq\r\ndef heap_queue_largest(nums,n):\r\n  largest_nums = hq.nlargest(n, nums)\r\n  return largest_nums' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是: {text} 你的代码应该通过这些测试:\n\n {test_list}  \n"
+                ),
+                dict(role="BOT", prompt="[BEGIN]\n"),
+            ], )),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer, max_out_len=512))
+
+mbpp_eval_cfg = dict(evaluator=dict(type=MBPPEvaluator), pred_role="BOT")
+
+mbpp_cn_datasets = [
+    dict(
+        type=MBPPDataset,
+        abbr='mbpp_cn',
+        path='./data/mbpp_cn/mbpp_cn.jsonl',
+        reader_cfg=mbpp_reader_cfg,
+        infer_cfg=mbpp_infer_cfg,
+        eval_cfg=mbpp_eval_cfg)
+]

configs/datasets/mbpp_cn/mbpp_cn_passk_gen_1d1481.py

@@ -0,0 +1,64 @@
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.datasets import MBPPDataset_V2, MBPPPassKEvaluator
+
+mbpp_reader_cfg = dict(
+    input_columns=['text', 'test_list'], output_column='test_column')
+
+mbpp_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            round=[
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个函数，从给定的两个元组列表中查找相似的元素。 你的代码应该通过这些测试：\n\n assert similar_elements((3, 4, 5, 6),(5, 7, 4, 10)) == (4, 5)\n assert similar_elements((1, 2, 3, 4),(5, 4, 3, 7)) == (3, 4) \n assert similar_elements((11, 12, 14, 13),(17, 15, 14, 13)) == (13, 14) \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'def similar_elements(test_tup1, test_tup2):\r\n  res = tuple(set(test_tup1) & set(test_tup2))\r\n  return (res)' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个 Python 函数来识别一个整数是否不是素数。 你的代码应该通过这些测试：\n\n assert is_not_prime(2) == False \n assert is_not_prime(10) == True \n assert is_not_prime(35) == True \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'import math\r\ndef is_not_prime(n):\r\n    result = False\r\n    for i in range(2,int(math.sqrt(n)) + 1):\r\n        if n % i == 0:\r\n            result = True\r\n    return result' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个函数，使用堆队列算法从给定的数字列表中查找最大整数。 你的代码应该通过这些测试：\n\n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],3)==[85, 75, 65] \n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],2)==[85, 75] \n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],5)==[85, 75, 65, 58, 35] \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'import heapq as hq\r\ndef heap_queue_largest(nums,n):\r\n  largest_nums = hq.nlargest(n, nums)\r\n  return largest_nums' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是: {text} 你的代码应该通过这些测试:\n\n {test_list}  \n"
+                ),
+                dict(role="BOT", prompt="[BEGIN]\n"),
+            ], )),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer, max_out_len=512))
+
+mbpp_eval_cfg = dict(evaluator=dict(type=MBPPPassKEvaluator), pred_role="BOT")
+
+mbpp_cn_datasets = [
+    dict(
+        type=MBPPDataset_V2,
+        abbr='mbpp_cn',
+        path='./data/mbpp_cn/mbpp_cn.jsonl',
+        reader_cfg=mbpp_reader_cfg,
+        infer_cfg=mbpp_infer_cfg,
+        eval_cfg=mbpp_eval_cfg)
+]

configs/datasets/mbpp_cn/mbpp_cn_repeat10_gen_1d1481.py

@@ -0,0 +1,65 @@
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.datasets import MBPPDataset_V2, MBPPPassKEvaluator
+
+mbpp_reader_cfg = dict(
+    input_columns=['text', 'test_list'], output_column='test_column')
+
+mbpp_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            round=[
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个函数，从给定的两个元组列表中查找相似的元素。 你的代码应该通过这些测试：\n\n assert similar_elements((3, 4, 5, 6),(5, 7, 4, 10)) == (4, 5)\n assert similar_elements((1, 2, 3, 4),(5, 4, 3, 7)) == (3, 4) \n assert similar_elements((11, 12, 14, 13),(17, 15, 14, 13)) == (13, 14) \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'def similar_elements(test_tup1, test_tup2):\r\n  res = tuple(set(test_tup1) & set(test_tup2))\r\n  return (res)' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个 Python 函数来识别一个整数是否不是素数。 你的代码应该通过这些测试：\n\n assert is_not_prime(2) == False \n assert is_not_prime(10) == True \n assert is_not_prime(35) == True \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'import math\r\ndef is_not_prime(n):\r\n    result = False\r\n    for i in range(2,int(math.sqrt(n)) + 1):\r\n        if n % i == 0:\r\n            result = True\r\n    return result' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是：编写一个函数，使用堆队列算法从给定的数字列表中查找最大整数。 你的代码应该通过这些测试：\n\n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],3)==[85, 75, 65] \n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],2)==[85, 75] \n assert heap_queue_largest( [25, 35, 22, 85, 14, 65, 75, 22, 58],5)==[85, 75, 65, 58, 35] \n"
+                ),
+                dict(
+                    role="BOT",
+                    prompt=
+                    "[BEGIN]\n 'import heapq as hq\r\ndef heap_queue_largest(nums,n):\r\n  largest_nums = hq.nlargest(n, nums)\r\n  return largest_nums' \n[DONE] \n\n "
+                ),
+                dict(
+                    role="HUMAN",
+                    prompt=
+                    "你是一名专业的 Python 程序员，你的任务是: {text} 你的代码应该通过这些测试:\n\n {test_list}  \n"
+                ),
+                dict(role="BOT", prompt="[BEGIN]\n"),
+            ], )),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer, max_out_len=512))
+
+mbpp_eval_cfg = dict(evaluator=dict(type=MBPPPassKEvaluator), pred_role="BOT")
+
+mbpp_cn_datasets = [
+    dict(
+        type=MBPPDataset_V2,
+        abbr='mbpp_cn_pass10',
+        path='./data/mbpp_cn/mbpp_cn.jsonl',
+        num_repeats=10,
+        reader_cfg=mbpp_reader_cfg,
+        infer_cfg=mbpp_infer_cfg,
+        eval_cfg=mbpp_eval_cfg)
+]

opencompass/datasets/mbpp.py

@@ -93,7 +93,10 @@ def load(path: str, num_repeats: int = 1):
 
         def processing_test(example):
             example['text'] = example.pop('prompt')
+            # used for prompt
             example['test_list'] = '\n'.join(example['test_list'])
+            # used for eval
+            example['test_list_2'] = example['test_list']
             example['test_column'] = dict(test_list_2=example['test_list'],
                                           task_id=example['task_id'])
             return example