Datasets:
refactor: script and readme
Browse files- README.md +34 -0
- selfie_and_video.py +18 -14
README.md
CHANGED
@@ -1,5 +1,39 @@
|
|
1 |
---
|
2 |
license: cc-by-nc-nd-4.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
4 |
# Selfies and video dataset
|
5 |
4000 people in this dataset. Each person took a selfie on a webcam, took a selfie on a mobile phone. In addition, people recorded video from the phone and from the webcam, on which they pronounced a given set of numbers.
|
|
|
1 |
---
|
2 |
license: cc-by-nc-nd-4.0
|
3 |
+
dataset_info:
|
4 |
+
features:
|
5 |
+
- name: photo_1
|
6 |
+
dtype: image
|
7 |
+
- name: photo_2
|
8 |
+
dtype: image
|
9 |
+
- name: video_3
|
10 |
+
dtype: string
|
11 |
+
- name: video_4
|
12 |
+
dtype: string
|
13 |
+
- name: photo_5
|
14 |
+
dtype: image
|
15 |
+
- name: photo_6
|
16 |
+
dtype: image
|
17 |
+
- name: video_7
|
18 |
+
dtype: string
|
19 |
+
- name: video_8
|
20 |
+
dtype: string
|
21 |
+
- name: set_id
|
22 |
+
dtype: string
|
23 |
+
- name: worker_id
|
24 |
+
dtype: string
|
25 |
+
- name: age
|
26 |
+
dtype: int8
|
27 |
+
- name: country
|
28 |
+
dtype: string
|
29 |
+
- name: gender
|
30 |
+
dtype: string
|
31 |
+
splits:
|
32 |
+
- name: train
|
33 |
+
num_bytes: 49771508
|
34 |
+
num_examples: 10
|
35 |
+
download_size: 829589647
|
36 |
+
dataset_size: 49771508
|
37 |
---
|
38 |
# Selfies and video dataset
|
39 |
4000 people in this dataset. Each person took a selfie on a webcam, took a selfie on a mobile phone. In addition, people recorded video from the phone and from the webcam, on which they pronounced a given set of numbers.
|
selfie_and_video.py
CHANGED
@@ -52,7 +52,7 @@ class SelfieAndVideo(datasets.GeneratorBasedBuilder):
|
|
52 |
)
|
53 |
|
54 |
def _split_generators(self, dl_manager):
|
55 |
-
images = dl_manager.download(f"{_DATA}
|
56 |
annotations = dl_manager.download(f"{_DATA}{_NAME}.csv")
|
57 |
images = dl_manager.iter_archive(images)
|
58 |
return [
|
@@ -64,7 +64,7 @@ class SelfieAndVideo(datasets.GeneratorBasedBuilder):
|
|
64 |
]
|
65 |
|
66 |
def _generate_examples(self, images, annotations):
|
67 |
-
annotations_df = pd.read_csv(annotations, sep='
|
68 |
images_data = pd.DataFrame(columns=['Link', 'Bytes'])
|
69 |
for idx, (image_path, image) in enumerate(images):
|
70 |
if image_path.lower().endswith('.jpg'):
|
@@ -73,28 +73,32 @@ class SelfieAndVideo(datasets.GeneratorBasedBuilder):
|
|
73 |
'Bytes': image.read()
|
74 |
}
|
75 |
|
76 |
-
annotations_df = pd.merge(annotations_df,
|
|
|
|
|
|
|
77 |
for idx, worker_id in enumerate(pd.unique(annotations_df['WorkerId'])):
|
78 |
annotation = annotations_df.loc[annotations_df['WorkerId'] ==
|
79 |
worker_id]
|
80 |
annotation = annotation.sort_values(['Link'])
|
81 |
data = {
|
82 |
-
(f'photo_{row[
|
83 |
({
|
84 |
-
'path': row[
|
85 |
-
'bytes': row[
|
86 |
-
} if row[
|
87 |
for row in annotation.itertuples()
|
88 |
}
|
|
|
89 |
|
90 |
-
age = annotation.loc[annotation['Link'].endswith(
|
91 |
'1.jpg')]['Age'].values[0]
|
92 |
-
country = annotation.loc[annotation['Link'].
|
93 |
-
|
94 |
-
gender = annotation.loc[annotation['Link'].
|
95 |
-
|
96 |
-
set_id = annotation.loc[annotation['Link'].
|
97 |
-
|
98 |
|
99 |
data['worker_id'] = worker_id
|
100 |
data['age'] = age
|
|
|
52 |
)
|
53 |
|
54 |
def _split_generators(self, dl_manager):
|
55 |
+
images = dl_manager.download(f"{_DATA}data.tar.gz")
|
56 |
annotations = dl_manager.download(f"{_DATA}{_NAME}.csv")
|
57 |
images = dl_manager.iter_archive(images)
|
58 |
return [
|
|
|
64 |
]
|
65 |
|
66 |
def _generate_examples(self, images, annotations):
|
67 |
+
annotations_df = pd.read_csv(annotations, sep=';')
|
68 |
images_data = pd.DataFrame(columns=['Link', 'Bytes'])
|
69 |
for idx, (image_path, image) in enumerate(images):
|
70 |
if image_path.lower().endswith('.jpg'):
|
|
|
73 |
'Bytes': image.read()
|
74 |
}
|
75 |
|
76 |
+
annotations_df = pd.merge(annotations_df,
|
77 |
+
images_data,
|
78 |
+
on=['Link'],
|
79 |
+
how='left')
|
80 |
for idx, worker_id in enumerate(pd.unique(annotations_df['WorkerId'])):
|
81 |
annotation = annotations_df.loc[annotations_df['WorkerId'] ==
|
82 |
worker_id]
|
83 |
annotation = annotation.sort_values(['Link'])
|
84 |
data = {
|
85 |
+
(f'photo_{row[7][37]}' if row[7].lower().endswith('.jpg') else f'video_{row[7][37]}'):
|
86 |
({
|
87 |
+
'path': row[7],
|
88 |
+
'bytes': row[8]
|
89 |
+
} if row[7].lower().endswith('.jpg') else row[7])
|
90 |
for row in annotation.itertuples()
|
91 |
}
|
92 |
+
print(annotation.head(8))
|
93 |
|
94 |
+
age = annotation.loc[annotation['Link'].str.lower().str.endswith(
|
95 |
'1.jpg')]['Age'].values[0]
|
96 |
+
country = annotation.loc[annotation['Link'].str.lower().str.
|
97 |
+
endswith('1.jpg')]['Country'].values[0]
|
98 |
+
gender = annotation.loc[annotation['Link'].str.lower().str.
|
99 |
+
endswith('1.jpg')]['Gender'].values[0]
|
100 |
+
set_id = annotation.loc[annotation['Link'].str.lower().str.
|
101 |
+
endswith('1.jpg')]['SetId'].values[0]
|
102 |
|
103 |
data['worker_id'] = worker_id
|
104 |
data['age'] = age
|