test_tasks.py 4.58 KB
Newer Older
1
import os
2
from itertools import islice
3

4
import datasets
5
import pytest
6

7
import lm_eval.tasks as tasks
8
from lm_eval.api.task import ConfigurableTask
9
from lm_eval.evaluator_utils import get_task_list
Leo Gao's avatar
Leo Gao committed
10

11
12
13
from .utils import new_tasks


14
datasets.config.HF_DATASETS_TRUST_REMOTE_CODE = True
15
os.environ["TOKENIZERS_PARALLELISM"] = "false"
16
task_manager = tasks.TaskManager()
baberabb's avatar
baberabb committed
17
# Default Task
Baber Abbasi's avatar
Baber Abbasi committed
18
TASKS = ["arc_easy"]
baberabb's avatar
baberabb committed
19
20
21
22
23
24
25


def task_class():
    global TASKS
    # CI: new_tasks checks if any modifications have been made
    task_classes = new_tasks()
    # Check if task_classes is empty
26
27
28
29
30
    task_classes = task_classes if task_classes else TASKS
    res = tasks.get_task_dict(task_classes, task_manager)
    res = [x.task for x in get_task_list(res)]

    return res
baberabb's avatar
baberabb committed
31
32
33


@pytest.fixture()
baberabb's avatar
baberabb committed
34
35
def limit() -> int:
    return 10
36
37
38


# Tests
Baber Abbasi's avatar
Baber Abbasi committed
39
@pytest.mark.parametrize("task_class", task_class(), ids=lambda x: f"{x.config.task}")
baberabb's avatar
baberabb committed
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
class TestNewTasks:
    def test_download(self, task_class: ConfigurableTask):
        task_class.download()
        assert task_class.dataset is not None

    def test_has_training_docs(self, task_class: ConfigurableTask):
        assert task_class.has_training_docs() in [True, False]

    def test_check_training_docs(self, task_class: ConfigurableTask):
        if task_class.has_training_docs():
            assert task_class._config["training_split"] is not None

    def test_has_validation_docs(self, task_class):
        assert task_class.has_validation_docs() in [True, False]

    def test_check_validation_docs(self, task_class):
        if task_class.has_validation_docs():
            assert task_class._config["validation_split"] is not None

    def test_has_test_docs(self, task_class):
        assert task_class.has_test_docs() in [True, False]

    def test_check_test_docs(self, task_class):
        task = task_class
        if task.has_test_docs():
            assert task._config["test_split"] is not None

    def test_should_decontaminate(self, task_class):
        task = task_class
        assert task.should_decontaminate() in [True, False]
        if task.should_decontaminate():
            assert task._config["doc_to_decontamination_query"] is not None

    def test_doc_to_text(self, task_class, limit):
        task = task_class
        arr = (
            list(islice(task.test_docs(), limit))
            if task.has_test_docs()
            else list(islice(task.validation_docs(), limit))
        )
        _array = [task.doc_to_text(doc) for doc in arr]
        # space convention; allow txt to have length 0 for perplexity-like tasks since the model tacks an <|endoftext|> on
82
83
84
85
86
87
88
        if not task.multiple_input:
            assert all(
                isinstance(x, str) and (x[-1] != " " if len(x) != 0 else True)
                for x in _array
            )
        else:
            pass
baberabb's avatar
baberabb committed
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110

    def test_create_choices(self, task_class, limit):
        task = task_class
        arr = (
            list(islice(task.test_docs(), limit))
            if task.has_test_docs()
            else list(islice(task.validation_docs(), limit))
        )
        if "multiple_choice" in task._config.output_type:
            _array = [task.doc_to_choice(doc) for doc in arr]
            assert all(isinstance(x, list) for x in _array)
            assert all(isinstance(x[0], str) for x in _array)

    def test_doc_to_target(self, task_class, limit):
        task = task_class
        arr = (
            list(islice(task.test_docs(), limit))
            if task.has_test_docs()
            else list(islice(task.validation_docs(), limit))
        )
        _array_target = [task.doc_to_target(doc) for doc in arr]
        if task._config.output_type == "multiple_choice":
111
112
113
114
115
            # TODO<baber>: label can be string or int; add better test conditions
            assert all(
                (isinstance(label, int) or isinstance(label, str))
                for label in _array_target
            )
baberabb's avatar
baberabb committed
116
117
118
119
120
121
122
123
124
125
126
127
128

    def test_build_all_requests(self, task_class, limit):
        task_class.build_all_requests(rank=1, limit=limit, world_size=1)
        assert task_class.instances is not None

    # ToDO: Add proper testing
    def test_construct_requests(self, task_class, limit):
        task = task_class
        arr = (
            list(islice(task.test_docs(), limit))
            if task.has_test_docs()
            else list(islice(task.validation_docs(), limit))
        )
129
130
131
132
133
134
135
        # ctx is "" for multiple input tasks
        requests = [
            task.construct_requests(
                doc=doc, ctx="" if task.multiple_input else task.doc_to_text(doc)
            )
            for doc in arr
        ]
baberabb's avatar
baberabb committed
136
        assert len(requests) == limit if limit else True