khaclinh commited on
Commit
d27f0f4
1 Parent(s): 38a1e07
.gitattributes CHANGED
@@ -49,3 +49,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
49
  *.jpg filter=lfs diff=lfs merge=lfs -text
50
  *.jpeg filter=lfs diff=lfs merge=lfs -text
51
  *.webp filter=lfs diff=lfs merge=lfs -text
 
 
 
49
  *.jpg filter=lfs diff=lfs merge=lfs -text
50
  *.jpeg filter=lfs diff=lfs merge=lfs -text
51
  *.webp filter=lfs diff=lfs merge=lfs -text
52
+ testdata.py filter=lfs diff=lfs merge=lfs -text
53
+ requirements.txt filter=lfs diff=lfs merge=lfs -text
data/annotations.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bfc9b871de6a1e37a0b275ccea678422e96fba72202ce1b8215cfebc69abbef0
3
- size 1034076
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aca1fed529a6f00598db63a1c685c3915d33d420430bad252a2822bac4b6609c
3
+ size 982986
data/{fisheye.zip → images.zip} RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6bea53ee6387cb672c3586993f3091b6ab18cc96e57809c3dfe9938629a5dbd4
3
- size 446670238
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:486d98266f56405840eda965c6d4e54008013e11456ed2e878abb9070cdc5286
3
+ size 4512838147
data/soiling_annotations.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2daec4c90f44c0f981c59d5ab9916ea6d14f95a938daffee1a863e37c69aea09
3
- size 1195308
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10731fc73eead247d280ce895d342f71b934eefdec1b7ab3622c92d7a3ddf516
3
+ size 1005826
requirements.txt CHANGED
@@ -1,7 +1,3 @@
1
- regex
2
- glob2
3
- tqdm
4
- pathlib
5
- collections2
6
- typing
7
- Pillow
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75d5618340023c3bab178f665513bee627f116ad66ba3a13963d7fe798330588
3
+ size 45
 
 
 
 
testdata.py CHANGED
@@ -1,151 +1,3 @@
1
- # coding=utf-8
2
- # Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
3
- #
4
- # Licensed under the Apache License, Version 2.0 (the "License");
5
- # you may not use this file except in compliance with the License.
6
- # You may obtain a copy of the License at
7
- #
8
- # http://www.apache.org/licenses/LICENSE-2.0
9
- #
10
- # Unless required by applicable law or agreed to in writing, software
11
- # distributed under the License is distributed on an "AS IS" BASIS,
12
- # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- # See the License for the specific language governing permissions and
14
- # limitations under the License.
15
- """PP4AV dataset."""
16
-
17
- import os
18
- from glob import glob
19
- from tqdm import tqdm
20
- from pathlib import Path
21
- from typing import List
22
- import re
23
- from collections import defaultdict
24
- import datasets
25
-
26
-
27
-
28
- _HOMEPAGE = "http://shuoyang1213.me/WIDERFACE/"
29
-
30
- _LICENSE = "Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)"
31
-
32
- _CITATION = """\
33
- @inproceedings{yang2016wider,
34
- Author = {Yang, Shuo and Luo, Ping and Loy, Chen Change and Tang, Xiaoou},
35
- Booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
36
- Title = {WIDER FACE: A Face Detection Benchmark},
37
- Year = {2016}}
38
- """
39
-
40
- _DESCRIPTION = """\
41
- WIDER FACE dataset is a face detection benchmark dataset, of which images are
42
- selected from the publicly available WIDER dataset. We choose 32,203 images and
43
- label 393,703 faces with a high degree of variability in scale, pose and
44
- occlusion as depicted in the sample images. WIDER FACE dataset is organized
45
- based on 61 event classes. For each event class, we randomly select 40%/10%/50%
46
- data as training, validation and testing sets. We adopt the same evaluation
47
- metric employed in the PASCAL VOC dataset. Similar to MALF and Caltech datasets,
48
- we do not release bounding box ground truth for the test images. Users are
49
- required to submit final prediction files, which we shall proceed to evaluate.
50
- """
51
-
52
-
53
- _REPO = "https://huggingface.co/datasets/khaclinh/testdata/resolve/main/data"
54
- _URLS = {
55
- "test": f"{_REPO}/fisheye.zip",
56
- "annot": f"{_REPO}/annotations.zip",
57
- }
58
-
59
- IMG_EXT = ['png', 'jpeg', 'jpg']
60
- _SUBREDDITS = ["zurich"]
61
-
62
- class TestDataConfig(datasets.BuilderConfig):
63
- """BuilderConfig for TestData."""
64
-
65
- def __init__(self, name, **kwargs):
66
- """BuilderConfig for TestData.
67
- Args:
68
- **kwargs: keyword arguments forwarded to super.
69
- """
70
- super(TestDataConfig, self).__init__(version=datasets.Version("1.0.0", ""), name=name, **kwargs)
71
-
72
- class TestData(datasets.GeneratorBasedBuilder):
73
- """WIDER FACE dataset."""
74
-
75
- BUILDER_CONFIGS = [
76
- TestDataConfig("fisheye"),
77
- ]
78
-
79
- BUILDER_CONFIGS += [TestDataConfig(subreddit) for subreddit in _SUBREDDITS]
80
-
81
- DEFAULT_CONFIG_NAME = "fisheye"
82
-
83
- VERSION = datasets.Version("1.0.0")
84
-
85
- def _info(self):
86
- return datasets.DatasetInfo(
87
- description=_DESCRIPTION,
88
- features=datasets.Features(
89
- {
90
- "image": datasets.Image(),
91
- "faces": datasets.Sequence(datasets.Sequence(datasets.Value("float32"), length=4)),
92
- "plates": datasets.Sequence(datasets.Sequence(datasets.Value("float32"), length=4)),
93
-
94
- }
95
- ),
96
- supervised_keys=None,
97
- homepage=_HOMEPAGE,
98
- license=_LICENSE,
99
- citation=_CITATION,
100
- )
101
-
102
- def _split_generators(self, dl_manager):
103
- data_dir = dl_manager.download_and_extract(_URLS)
104
- return [
105
- datasets.SplitGenerator(
106
- name=datasets.Split.TEST,
107
- gen_kwargs={
108
- "name": self.config.name,
109
- "data_dir": data_dir["test"],
110
- "annot_dir": data_dir["annot"],
111
- },
112
- ),
113
- ]
114
-
115
- def _generate_examples(self, name, data_dir, annot_dir):
116
-
117
- image_dir = os.path.join(data_dir, name)
118
- annotation_dir = os.path.join(annot_dir, name)
119
- files = []
120
-
121
- idx = 0
122
- for i_file in glob(os.path.join(image_dir, "*.png")):
123
- plates = []
124
- faces = []
125
-
126
- img_relative_file = os.path.relpath(i_file, image_dir)
127
- gt_relative_path = img_relative_file.replace(".png", ".txt")
128
-
129
- gt_path = os.path.join(annotation_dir, gt_relative_path)
130
-
131
- annotation = defaultdict(list)
132
- with open(gt_path, "r", encoding="utf-8") as f:
133
- line = f.readline().strip()
134
- while line:
135
- assert re.match(r"^\d( [\d\.]+){4,5}$", line), "Incorrect line: %s" % line
136
- cls, cx, cy, w, h = line.split()[:5]
137
- cls, cx, cy, w, h = int(cls), float(cx), float(cy), float(w), float(h)
138
- x1, y1, x2, y2 = cx - w / 2, cy - h / 2, cx + w / 2, cy + h / 2
139
- annotation[cls].append([x1, y1, x2, y2])
140
- line = f.readline().strip()
141
-
142
- for cls, bboxes in annotation.items():
143
- for x1, y1, x2, y2 in bboxes:
144
- if cls == 0:
145
- faces.append([x1, y1, x2, y2])
146
- else:
147
- plates.append([x1, y1, x2, y2])
148
-
149
- yield idx, {"image": i_file, "faces": faces, "plates": plates}
150
-
151
- idx += 1
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a3c670d21b2609d4427111d778dfec3438529f196a2287b778541809e357c54
3
+ size 5682