File size: 1,070 Bytes
01f199c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
from .Dataset import Dataset
from evaluations.func_evaluate import evaluate_functional_correctness, evaluate_io
from constants.paths import *


class HumanDataset(Dataset):
    def __init__(
        self,
        path: str = HUMAN_WST_DATA_PATH,
    ):
        super().__init__(path)
        self.id_key = "task_id"

    def evaluate(
        self,
        item: dict,
        cur_imp: str,
        language: str,
    ):
        result = evaluate_functional_correctness(
            problem=item,
            completion=cur_imp
        )
        return result == "passed"

    def evaluate_sample_io(
        self,
        item: dict,
        cur_imp: str,
        language: str,
    ):
        
        return evaluate_io(
            sample_io=item["sample_io"],
            completion=cur_imp,            
        )


    @staticmethod
    def get_prompt(item):
        if "prompt" in item:
            return f"{item['prompt']}"
        elif "text" in item:
            return f"{item['text']}"
        else:
            raise Exception("No prompt or text in item")