File size: 2,510 Bytes
da3e9e3 |
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 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 |
from scripts.generate import MetashiftFactory, MetashiftData, get_dataset_name
import json
from omegaconf import OmegaConf
import random
import pytest
CONFIG_PATH = "configs/generate.yaml"
CANDIDATE_SUBSETS_PATH = "scripts/artifacts/csp.pkl"
def test_consistent_generate():
config = OmegaConf.load(CONFIG_PATH)
metashift_factory = MetashiftFactory(
full_candidate_subsets_path=CANDIDATE_SUBSETS_PATH,
visual_genome_images_dir=".",
)
info: dict[str, MetashiftData] = dict()
for task_config in config.tasks:
for experiment_config in task_config.experiments:
data = metashift_factory.create(
seed=task_config.seed,
selected_classes=task_config.selected_classes,
spurious_class=experiment_config.spurious_class,
train_spurious_context=experiment_config.train_context,
test_spurious_context=experiment_config.test_context,
num_test_images_per_class=task_config.num_images_per_class_test,
num_train_images_per_class=task_config.num_images_per_class_train,
)
dataset_name = get_dataset_name(task_config.name, experiment_config.name)
assert dataset_name not in info
info[dataset_name] = data
random.seed(2)
unique_ids = metashift_factory._get_unique_ids_from_info(info)
random.seed(10000)
unique_ids_2 = metashift_factory._get_unique_ids_from_info(info)
assert unique_ids == unique_ids_2
@pytest.mark.parametrize(
"scenario_path",
[
"task_1_bed_cat_dog.json",
"task_1_bed_dog_cat.json",
"task_2_table_books_cat.json",
"task_2_table_books_dog.json",
"task_2_table_cat_dog.json",
"task_2_table_dog_cat.json",
],
)
def test_unique_train_test_ids(scenario_path: str):
""" Test that the train and test sets have unique ids."""
with open(f"scenarios/{scenario_path}", "r") as f:
scenario = json.load(f)
train_unique_ids = set()
test_unique_ids = set()
for train_ids in scenario["data_splits"]["train"].values():
for train_id in train_ids:
train_unique_ids.add(train_id)
for test_ids in scenario["data_splits"]["test"].values():
for test_id in test_ids:
test_unique_ids.add(test_id)
intersection = train_unique_ids.intersection(test_unique_ids)
assert len(intersection) == 0, len(intersection) |