Hugging Face's logo

Spaces:
Gotech
/
cad-benchmark
Running on Zero

App Files Community
cad-benchmark
File size: 4,974 Bytes
99da741
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
import pytest
from pathlib import Path
import numpy as np
import nibabel as nib

from cad_benchmark_space.validator import validate_submission


@pytest.fixture
def mock_gt():
    return {
        "canonical_orientation": "RAS+",
        "cases": {
            "case001": {
                "labels": {"2c": 1, "2d": 0},
                "shape": (32, 64, 64),
                "spacing": (1.5, 0.75, 0.75),
                "orientation": ("R", "A", "S"),
                "findings": {
                    "case001_f0": {"class": "2c", "mask_path": "masks/case001_f0.nii.gz"}
                },
                "gt_masks": {"2c": np.zeros((32, 64, 64), dtype=bool), "2d": np.zeros((32, 64, 64), dtype=bool)},
            },
            "case002": {
                "labels": {"2c": 0, "2d": 1},
                "shape": (32, 64, 64),
                "spacing": (1.5, 0.75, 0.75),
                "orientation": ("R", "A", "S"),
                "findings": {
                    "case002_f0": {"class": "2d", "mask_path": "masks/case002_f0.nii.gz"}
                },
                "gt_masks": {"2c": np.zeros((32, 64, 64), dtype=bool), "2d": np.zeros((32, 64, 64), dtype=bool)},
            },
        },
    }


def test_reject_invalid_model_type(mock_gt):
    manifest = {"model_name": "TestModel", "model_type": "invalid_type"}
    res = validate_submission(manifest, mock_gt)
    assert not res.is_valid
    assert any("model_type" in err for err in res.errors)


def test_classification_reject_floats(mock_gt):
    manifest = {
        "model_name": "TestModel",
        "model_type": "classification",
        "predictions": [
            {"scan_id": "case001", "2c": 0.85, "2d": 0}
        ],
    }
    res = validate_submission(manifest, mock_gt)
    assert not res.is_valid
    assert any("must be int 0 or 1" in err for err in res.errors)


def test_classification_reject_unknown_scan_id(mock_gt):
    manifest = {
        "model_name": "TestModel",
        "model_type": "classification",
        "predictions": [
            {"scan_id": "case999", "2c": 1, "2d": 0}
        ],
    }
    res = validate_submission(manifest, mock_gt)
    assert not res.is_valid
    assert any("case999" in err for err in res.errors)


def test_classification_reject_duplicates(mock_gt):
    manifest = {
        "model_name": "TestModel",
        "model_type": "classification",
        "predictions": [
            {"scan_id": "case001", "2c": 1, "2d": 0},
            {"scan_id": "case001", "2c": 0, "2d": 1},
        ],
    }
    res = validate_submission(manifest, mock_gt)
    assert not res.is_valid
    assert any("duplicate scan_id" in err for err in res.errors)


def test_classification_valid(mock_gt):
    manifest = {
        "model_name": "TestModel",
        "model_type": "classification",
        "predictions": [
            {"scan_id": "case001", "2c": 1, "2d": 0},
            {"scan_id": "case002", "2c": 0, "2d": 1},
        ],
    }
    res = validate_submission(manifest, mock_gt)
    assert res.is_valid
    assert len(res.classification_samples) == 2


def test_segmentation_validation(mock_gt, tmp_path):
    shape = (32, 64, 64)
    spacing = (1.5, 0.75, 0.75)
    affine = np.diag([spacing[0], spacing[1], spacing[2], 1.0])
    mask_data = np.zeros(shape, dtype=np.uint8)
    mask_data[10:15, 20:30, 20:30] = 1

    mask_file = tmp_path / "case001_f0.nii.gz"
    nii = nib.Nifti1Image(mask_data, affine)
    nib.save(nii, str(mask_file))

    manifest = {
        "model_name": "SegModel",
        "model_type": "segmentation",
        "predictions": [
            {
                "scan_id": "case001",
                "finding_id": "case001_f0",
                "class": "2c",
                "mask_path": "case001_f0.nii.gz",
                "is_soft_mask": False,
            }
        ],
    }

    res = validate_submission(manifest, mock_gt, files_dir=tmp_path)
    assert res.is_valid
    assert len(res.localization_samples) == 1


def test_segmentation_reject_shape_mismatch(mock_gt, tmp_path):
    wrong_shape = (20, 64, 64)
    spacing = (1.5, 0.75, 0.75)
    affine = np.diag([spacing[0], spacing[1], spacing[2], 1.0])
    mask_data = np.zeros(wrong_shape, dtype=np.uint8)

    mask_file = tmp_path / "case001_wrong.nii.gz"
    nii = nib.Nifti1Image(mask_data, affine)
    nib.save(nii, str(mask_file))

    manifest = {
        "model_name": "SegModel",
        "model_type": "segmentation",
        "predictions": [
            {
                "scan_id": "case001",
                "finding_id": "case001_f0",
                "class": "2c",
                "mask_path": "case001_wrong.nii.gz",
            }
        ],
    }

    res = validate_submission(manifest, mock_gt, files_dir=tmp_path)
    assert not res.is_valid
    assert any("mask shape" in err for err in res.errors)