Vadzim Kashko commited on
Commit
46a9e6b
1 Parent(s): 4add8b3

feat: script

Browse files
Files changed (1) hide show
  1. facial-emotion-recognition-dataset.py +183 -0
facial-emotion-recognition-dataset.py ADDED
@@ -0,0 +1,183 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import json
2
+ from pathlib import Path
3
+
4
+ import datasets
5
+ import numpy as np
6
+ import pandas as pd
7
+ import PIL.Image
8
+ import PIL.ImageOps
9
+
10
+ _CITATION = """\
11
+ @InProceedings{huggingface:dataset,
12
+ title = {body-measurements-dataset},
13
+ author = {TrainingDataPro},
14
+ year = {2023}
15
+ }
16
+ """
17
+
18
+ _DESCRIPTION = """\
19
+ The dataset consists of a compilation of people's photos along with their
20
+ corresponding body measurements. It is designed to provide information and
21
+ insights into the physical appearances and body characteristics of individuals.
22
+ The dataset includes a diverse range of subjects representing different age
23
+ groups, genders, and ethnicities.
24
+
25
+ The photos are captured in a standardized manner, depicting individuals in a
26
+ front and side positions.
27
+ The images aim to capture the subjects' physical appearance using appropriate
28
+ lighting and angles that showcase their body proportions accurately.
29
+
30
+ The dataset serves various purposes, including:
31
+ - research projects
32
+ - body measurement analysis
33
+ - fashion or apparel industry applications
34
+ - fitness and wellness studies
35
+ - anthropometric studies for ergonomic design in various fields
36
+ """
37
+ _NAME = 'body-measurements-dataset'
38
+
39
+ _HOMEPAGE = f"https://huggingface.co/datasets/TrainingDataPro/{_NAME}"
40
+
41
+ _LICENSE = "cc-by-nc-nd-4.0"
42
+
43
+ _DATA = f"https://huggingface.co/datasets/TrainingDataPro/{_NAME}/resolve/main/data/"
44
+
45
+
46
+ class BodyMeasurementsDataset(datasets.GeneratorBasedBuilder):
47
+
48
+ def _info(self):
49
+ return datasets.DatasetInfo(
50
+ description=_DESCRIPTION,
51
+ features=datasets.Features({
52
+ 'front_img': datasets.Image(),
53
+ 'selfie_img': datasets.Image(),
54
+ 'side_img': datasets.Image(),
55
+ "arm_circumference_cm": datasets.Value('string'),
56
+ "arm_length_cm": datasets.Value('string'),
57
+ "back_build_cm": datasets.Value('string'),
58
+ "calf_circumference_cm": datasets.Value('string'),
59
+ "chest_circumference_cm": datasets.Value('string'),
60
+ "crotch_height_cm": datasets.Value('string'),
61
+ "front_build_cm": datasets.Value('string'),
62
+ "hips_circumference_cm": datasets.Value('string'),
63
+ "leg_length_cm": datasets.Value('string'),
64
+ "neck_circumference_cm": datasets.Value('string'),
65
+ "neck_pelvis_length_front_cm": datasets.Value('string'),
66
+ "neck_waist_length_back_cm": datasets.Value('string'),
67
+ "neck_waist_length_front_cm": datasets.Value('string'),
68
+ "pelvis_circumference_cm": datasets.Value('string'),
69
+ "shoulder_length_cm": datasets.Value('string'),
70
+ "shoulder_width_cm": datasets.Value('string'),
71
+ "thigh_circumference_cm": datasets.Value('string'),
72
+ "under_chest_circumference_cm": datasets.Value('string'),
73
+ "upper_arm_length_cm": datasets.Value('string'),
74
+ "waist_circumference_cm": datasets.Value('string'),
75
+ "height": datasets.Value('string'),
76
+ "weight": datasets.Value('string'),
77
+ "age": datasets.Value('string'),
78
+ "gender": datasets.Value('string'),
79
+ "race": datasets.Value('string'),
80
+ "profession": datasets.Value('string'),
81
+ "arm_circumference": datasets.Image(),
82
+ "arm_length": datasets.Image(),
83
+ "back_build": datasets.Image(),
84
+ "calf_circumference": datasets.Image(),
85
+ "chest_circumference": datasets.Image(),
86
+ "crotch_height": datasets.Image(),
87
+ "front_build": datasets.Image(),
88
+ "hips_circumference": datasets.Image(),
89
+ "leg_length": datasets.Image(),
90
+ "neck_circumference": datasets.Image(),
91
+ "neck_pelvis_length_front": datasets.Image(),
92
+ "neck_waist_length_back": datasets.Image(),
93
+ "neck_waist_length_front": datasets.Image(),
94
+ "pelvis_circumference": datasets.Image(),
95
+ "shoulder_length": datasets.Image(),
96
+ "shoulder_width": datasets.Image(),
97
+ "thigh_circumference": datasets.Image(),
98
+ "under_chest_circumference": datasets.Image(),
99
+ "upper_arm_length": datasets.Image(),
100
+ "waist_circumference": datasets.Image()
101
+ }),
102
+ supervised_keys=None,
103
+ homepage=_HOMEPAGE,
104
+ citation=_CITATION,
105
+ license=_LICENSE)
106
+
107
+ def _split_generators(self, dl_manager):
108
+ files = dl_manager.download_and_extract(f"{_DATA}files.zip")
109
+ proofs = dl_manager.download_and_extract(f"{_DATA}proofs.zip")
110
+ annotations = dl_manager.download(f"{_DATA}{_NAME}.csv")
111
+ files = dl_manager.iter_files(files)
112
+ proofs = dl_manager.iter_files(proofs)
113
+ return [
114
+ datasets.SplitGenerator(name=datasets.Split.TRAIN,
115
+ gen_kwargs={
116
+ "files": files,
117
+ 'proofs': proofs,
118
+ 'annotations': annotations
119
+ }),
120
+ ]
121
+
122
+ def _generate_examples(self, files, proofs, annotations):
123
+ files = list(files)
124
+ files = [files[i:i + 4] for i in range(0, len(files), 4)]
125
+ proofs = list(proofs)
126
+ proofs = [proofs[i:i + 20] for i in range(0, len(proofs), 20)]
127
+
128
+ for idx, (files_dir, proofs_dir) in enumerate(zip(files, proofs)):
129
+ data = {}
130
+ for file in files_dir:
131
+ if 'front_img' in file:
132
+ data['front_img'] = file
133
+ elif 'selfie_img' in file:
134
+ data['selfie_img'] = file
135
+ elif 'side_img' in file:
136
+ data['side_img'] = file
137
+ elif 'measurements' in file:
138
+ with open(file) as f:
139
+ data.update(json.load(f))
140
+
141
+ for proof in proofs_dir:
142
+ if "arm_circumference" in proof:
143
+ data['arm_circumference'] = proof
144
+ elif 'upper_arm_length' in proof:
145
+ data['upper_arm_length'] = proof
146
+ elif 'arm_length' in proof:
147
+ data['arm_length'] = proof
148
+ elif 'back_build' in proof:
149
+ data['back_build'] = proof
150
+ elif 'calf_circumference' in proof:
151
+ data['calf_circumference'] = proof
152
+ elif 'under_chest_circumference' in proof:
153
+ data['under_chest_circumference'] = proof
154
+ elif 'chest_circumference' in proof:
155
+ data['chest_circumference'] = proof
156
+ elif 'crotch_height' in proof:
157
+ data['crotch_height'] = proof
158
+ elif 'front_build' in proof:
159
+ data['front_build'] = proof
160
+ elif 'hips_circumference' in proof:
161
+ data['hips_circumference'] = proof
162
+ elif 'leg_length' in proof:
163
+ data['leg_length'] = proof
164
+ elif 'neck_circumference' in proof:
165
+ data['neck_circumference'] = proof
166
+ elif 'neck_pelvis_length_front' in proof:
167
+ data['neck_pelvis_length_front'] = proof
168
+ elif 'neck_waist_length_back' in proof:
169
+ data['neck_waist_length_back'] = proof
170
+ elif 'neck_waist_length_front' in proof:
171
+ data['neck_waist_length_front'] = proof
172
+ elif 'pelvis_circumference' in proof:
173
+ data['pelvis_circumference'] = proof
174
+ elif 'shoulder_length' in proof:
175
+ data['shoulder_length'] = proof
176
+ elif 'shoulder_width' in proof:
177
+ data['shoulder_width'] = proof
178
+ elif 'thigh_circumference' in proof:
179
+ data['thigh_circumference'] = proof
180
+ elif 'waist_circumference' in proof:
181
+ data['waist_circumference'] = proof
182
+
183
+ yield idx, data