Datasets:
Commit
•
8e33d10
1
Parent(s):
627ad4b
add coco config
Browse files- dataset_infos.json +1 -1
- yalt_ai_tabular_dataset.py +155 -27
dataset_infos.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"default": {"description": "TODO", "citation": " @dataset{clerice_thibault_2022_6827706,\n author = {Cl\u00e9rice, Thibault},\n title = {YALTAi: Tabular Dataset},\n month = jul,\n year = 2022,\n publisher = {Zenodo},\n version = {1.0.0},\n doi = {10.5281/zenodo.6827706},\n url = {https://doi.org/10.5281/zenodo.6827706}\n}\n", "homepage": "https://doi.org/10.5281/zenodo.6827706", "license": "Creative Commons Attribution 4.0 International", "features": {"image": {"decode": true, "id": null, "_type": "Image"}, "objects": {"feature": {"label": {"num_classes": 4, "names": ["Header", "Col", "Marginal", "text"], "id": null, "_type": "ClassLabel"}, "bbox": {"feature": {"dtype": "int32", "id": null, "_type": "Value"}, "length": 4, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "yalt_ai_tabular_dataset", "config_name": "default", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 60704, "num_examples": 196, "dataset_name": "yalt_ai_tabular_dataset"}, "validation": {"name": "validation", "num_bytes": 7537, "num_examples": 22, "dataset_name": "yalt_ai_tabular_dataset"}, "test": {"name": "test", "num_bytes": 47159, "num_examples": 135, "dataset_name": "yalt_ai_tabular_dataset"}}, "download_checksums": {"https://zenodo.org/record/6827706/files/yaltai-table.zip?download=1": {"num_bytes": 376190064, "checksum": "5b312faf097939302fb98ab0a8b35c007962d88978ea9dc28d2f560b89dc0657"}}, "download_size": 376190064, "post_processing_size": null, "dataset_size": 115400, "size_in_bytes": 376305464}}
|
|
|
1 |
+
{"default": {"description": "TODO", "citation": " @dataset{clerice_thibault_2022_6827706,\n author = {Cl\u00e9rice, Thibault},\n title = {YALTAi: Tabular Dataset},\n month = jul,\n year = 2022,\n publisher = {Zenodo},\n version = {1.0.0},\n doi = {10.5281/zenodo.6827706},\n url = {https://doi.org/10.5281/zenodo.6827706}\n}\n", "homepage": "https://doi.org/10.5281/zenodo.6827706", "license": "Creative Commons Attribution 4.0 International", "features": {"image": {"decode": true, "id": null, "_type": "Image"}, "objects": {"feature": {"label": {"num_classes": 4, "names": ["Header", "Col", "Marginal", "text"], "id": null, "_type": "ClassLabel"}, "bbox": {"feature": {"dtype": "int32", "id": null, "_type": "Value"}, "length": 4, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "yalt_ai_tabular_dataset", "config_name": "default", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 60704, "num_examples": 196, "dataset_name": "yalt_ai_tabular_dataset"}, "validation": {"name": "validation", "num_bytes": 7537, "num_examples": 22, "dataset_name": "yalt_ai_tabular_dataset"}, "test": {"name": "test", "num_bytes": 47159, "num_examples": 135, "dataset_name": "yalt_ai_tabular_dataset"}}, "download_checksums": {"https://zenodo.org/record/6827706/files/yaltai-table.zip?download=1": {"num_bytes": 376190064, "checksum": "5b312faf097939302fb98ab0a8b35c007962d88978ea9dc28d2f560b89dc0657"}}, "download_size": 376190064, "post_processing_size": null, "dataset_size": 115400, "size_in_bytes": 376305464}, "YOLO": {"description": "TODO", "citation": " @dataset{clerice_thibault_2022_6827706,\n author = {Cl\u00e9rice, Thibault},\n title = {YALTAi: Tabular Dataset},\n month = jul,\n year = 2022,\n publisher = {Zenodo},\n version = {1.0.0},\n doi = {10.5281/zenodo.6827706},\n url = {https://doi.org/10.5281/zenodo.6827706}\n}\n", "homepage": "https://doi.org/10.5281/zenodo.6827706", "license": "Creative Commons Attribution 4.0 International", "features": {"image": {"decode": true, "id": null, "_type": "Image"}, "objects": {"feature": {"label": {"num_classes": 4, "names": ["Header", "Col", "Marginal", "text"], "id": null, "_type": "ClassLabel"}, "bbox": {"feature": {"dtype": "int32", "id": null, "_type": "Value"}, "length": 4, "id": null, "_type": "Sequence"}}, "length": -1, "id": null, "_type": "Sequence"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "yalt_ai_tabular_dataset", "config_name": "YOLO", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 60704, "num_examples": 196, "dataset_name": "yalt_ai_tabular_dataset"}, "validation": {"name": "validation", "num_bytes": 7537, "num_examples": 22, "dataset_name": "yalt_ai_tabular_dataset"}, "test": {"name": "test", "num_bytes": 47159, "num_examples": 135, "dataset_name": "yalt_ai_tabular_dataset"}}, "download_checksums": {"https://zenodo.org/record/6827706/files/yaltai-table.zip?download=1": {"num_bytes": 376190064, "checksum": "5b312faf097939302fb98ab0a8b35c007962d88978ea9dc28d2f560b89dc0657"}}, "download_size": 376190064, "post_processing_size": null, "dataset_size": 115400, "size_in_bytes": 376305464}, "COCO": {"description": "TODO", "citation": " @dataset{clerice_thibault_2022_6827706,\n author = {Cl\u00e9rice, Thibault},\n title = {YALTAi: Tabular Dataset},\n month = jul,\n year = 2022,\n publisher = {Zenodo},\n version = {1.0.0},\n doi = {10.5281/zenodo.6827706},\n url = {https://doi.org/10.5281/zenodo.6827706}\n}\n", "homepage": "https://doi.org/10.5281/zenodo.6827706", "license": "Creative Commons Attribution 4.0 International", "features": {"image_id": {"dtype": "int64", "id": null, "_type": "Value"}, "image": {"decode": true, "id": null, "_type": "Image"}, "width": {"dtype": "int32", "id": null, "_type": "Value"}, "height": {"dtype": "int32", "id": null, "_type": "Value"}, "objects": [{"category_id": {"num_classes": 4, "names": ["Header", "Col", "Marginal", "text"], "id": null, "_type": "ClassLabel"}, "image_id": {"dtype": "string", "id": null, "_type": "Value"}, "id": {"dtype": "int64", "id": null, "_type": "Value"}, "area": {"dtype": "int64", "id": null, "_type": "Value"}, "bbox": {"feature": {"dtype": "float32", "id": null, "_type": "Value"}, "length": 4, "id": null, "_type": "Sequence"}, "segmentation": [[{"dtype": "float32", "id": null, "_type": "Value"}]], "iscrowd": {"dtype": "bool", "id": null, "_type": "Value"}}]}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "yalt_ai_tabular_dataset", "config_name": "COCO", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 87171, "num_examples": 196, "dataset_name": "yalt_ai_tabular_dataset"}, "validation": {"name": "validation", "num_bytes": 11225, "num_examples": 22, "dataset_name": "yalt_ai_tabular_dataset"}, "test": {"name": "test", "num_bytes": 71491, "num_examples": 135, "dataset_name": "yalt_ai_tabular_dataset"}}, "download_checksums": {"https://zenodo.org/record/6827706/files/yaltai-table.zip?download=1": {"num_bytes": 376190064, "checksum": "5b312faf097939302fb98ab0a8b35c007962d88978ea9dc28d2f560b89dc0657"}}, "download_size": 376190064, "post_processing_size": null, "dataset_size": 169887, "size_in_bytes": 376359951}}
|
yalt_ai_tabular_dataset.py
CHANGED
@@ -16,6 +16,7 @@
|
|
16 |
|
17 |
import os
|
18 |
from glob import glob
|
|
|
19 |
|
20 |
import datasets
|
21 |
from PIL import Image
|
@@ -44,15 +45,49 @@ _URL = "https://zenodo.org/record/6827706/files/yaltai-table.zip?download=1"
|
|
44 |
_CATEGORIES = ["Header", "Col", "Marginal", "text"]
|
45 |
|
46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
47 |
class YaltAiTabularDataset(datasets.GeneratorBasedBuilder):
|
48 |
"""Object Detection for historic manuscripts"""
|
49 |
|
50 |
-
|
|
|
|
|
|
|
51 |
|
52 |
def _info(self):
|
53 |
-
|
54 |
-
features=datasets.Features(
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
{
|
|
|
56 |
"image": datasets.Image(),
|
57 |
"objects": datasets.Sequence(
|
58 |
{
|
@@ -63,7 +98,9 @@ class YaltAiTabularDataset(datasets.GeneratorBasedBuilder):
|
|
63 |
}
|
64 |
),
|
65 |
}
|
66 |
-
)
|
|
|
|
|
67 |
supervised_keys=None,
|
68 |
description=_DESCRIPTION,
|
69 |
homepage=_HOMEPAGE,
|
@@ -93,31 +130,122 @@ class YaltAiTabularDataset(datasets.GeneratorBasedBuilder):
|
|
93 |
]
|
94 |
|
95 |
def _generate_examples(self, data_dir):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
96 |
image_dir = os.path.join(data_dir, "images")
|
97 |
label_dir = os.path.join(data_dir, "labels")
|
98 |
image_paths = sorted(glob(f"{image_dir}/*.jpg"))
|
99 |
label_paths = sorted(glob(f"{label_dir}/*.txt"))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
100 |
|
101 |
-
|
102 |
-
|
103 |
-
|
104 |
-
|
105 |
-
with open(label_path, "r") as f:
|
106 |
-
lines = f.readlines()
|
107 |
-
|
108 |
-
objects = []
|
109 |
-
for line in lines:
|
110 |
-
line = line.strip().split()
|
111 |
-
bbox_class = int(line[0])
|
112 |
-
bbox_xcenter = int(float(line[1]) * width)
|
113 |
-
bbox_ycenter = int(float(line[2]) * height)
|
114 |
-
bbox_width = int(float(line[3]) * width)
|
115 |
-
bbox_height = int(float(line[4]) * height)
|
116 |
-
objects.append(
|
117 |
-
{
|
118 |
-
"label": bbox_class,
|
119 |
-
"bbox": [bbox_xcenter, bbox_ycenter, bbox_width, bbox_height],
|
120 |
-
}
|
121 |
-
)
|
122 |
-
|
123 |
-
yield idx, {"image": image_path, "objects": objects}
|
|
|
16 |
|
17 |
import os
|
18 |
from glob import glob
|
19 |
+
from re import L
|
20 |
|
21 |
import datasets
|
22 |
from PIL import Image
|
|
|
45 |
_CATEGORIES = ["Header", "Col", "Marginal", "text"]
|
46 |
|
47 |
|
48 |
+
class YaltAiTabularDatasetConfig(datasets.BuilderConfig):
|
49 |
+
"""BuilderConfig for YaltAiTabularDataset."""
|
50 |
+
|
51 |
+
def __init__(self, name, **kwargs):
|
52 |
+
"""BuilderConfig for YaltAiTabularDataset."""
|
53 |
+
super(YaltAiTabularDatasetConfig, self).__init__(
|
54 |
+
version=datasets.Version("1.0.0"), name=name, description=None, **kwargs
|
55 |
+
)
|
56 |
+
|
57 |
+
|
58 |
class YaltAiTabularDataset(datasets.GeneratorBasedBuilder):
|
59 |
"""Object Detection for historic manuscripts"""
|
60 |
|
61 |
+
BUILDER_CONFIGS = [
|
62 |
+
YaltAiTabularDatasetConfig("YOLO"),
|
63 |
+
YaltAiTabularDatasetConfig("COCO"),
|
64 |
+
]
|
65 |
|
66 |
def _info(self):
|
67 |
+
if self.config.name == "COCO":
|
68 |
+
features = datasets.Features(
|
69 |
+
{
|
70 |
+
"image_id": datasets.Value("int64"),
|
71 |
+
"image": datasets.Image(),
|
72 |
+
"width": datasets.Value("int32"),
|
73 |
+
"height": datasets.Value("int32"),
|
74 |
+
# "url": datasets.Value("string"),
|
75 |
+
}
|
76 |
+
)
|
77 |
+
object_dict = {
|
78 |
+
"category_id": datasets.ClassLabel(names=_CATEGORIES),
|
79 |
+
"image_id": datasets.Value("string"),
|
80 |
+
"id": datasets.Value("int64"),
|
81 |
+
"area": datasets.Value("int64"),
|
82 |
+
"bbox": datasets.Sequence(datasets.Value("float32"), length=4),
|
83 |
+
"segmentation": [[datasets.Value("float32")]],
|
84 |
+
"iscrowd": datasets.Value("bool"),
|
85 |
+
}
|
86 |
+
features["objects"] = [object_dict]
|
87 |
+
if self.config.name == "YOLO":
|
88 |
+
features = datasets.Features(
|
89 |
{
|
90 |
+
# "image_id": datasets.Value("int32"),
|
91 |
"image": datasets.Image(),
|
92 |
"objects": datasets.Sequence(
|
93 |
{
|
|
|
98 |
}
|
99 |
),
|
100 |
}
|
101 |
+
)
|
102 |
+
return datasets.DatasetInfo(
|
103 |
+
features=features,
|
104 |
supervised_keys=None,
|
105 |
description=_DESCRIPTION,
|
106 |
homepage=_HOMEPAGE,
|
|
|
130 |
]
|
131 |
|
132 |
def _generate_examples(self, data_dir):
|
133 |
+
def create_annotation_from_yolo_format(
|
134 |
+
min_x,
|
135 |
+
min_y,
|
136 |
+
width,
|
137 |
+
height,
|
138 |
+
image_id,
|
139 |
+
category_id,
|
140 |
+
annotation_id,
|
141 |
+
segmentation=False,
|
142 |
+
):
|
143 |
+
bbox = (float(min_x), float(min_y), float(width), float(height))
|
144 |
+
area = width * height
|
145 |
+
max_x = min_x + width
|
146 |
+
max_y = min_y + height
|
147 |
+
if segmentation:
|
148 |
+
seg = [[min_x, min_y, max_x, min_y, max_x, max_y, min_x, max_y]]
|
149 |
+
else:
|
150 |
+
seg = []
|
151 |
+
return {
|
152 |
+
"id": annotation_id,
|
153 |
+
"image_id": image_id,
|
154 |
+
"bbox": bbox,
|
155 |
+
"area": area,
|
156 |
+
"iscrowd": 0,
|
157 |
+
"category_id": category_id,
|
158 |
+
"segmentation": seg,
|
159 |
+
}
|
160 |
+
|
161 |
image_dir = os.path.join(data_dir, "images")
|
162 |
label_dir = os.path.join(data_dir, "labels")
|
163 |
image_paths = sorted(glob(f"{image_dir}/*.jpg"))
|
164 |
label_paths = sorted(glob(f"{label_dir}/*.txt"))
|
165 |
+
if self.config.name == "COCO":
|
166 |
+
for idx, (image_path, label_path) in enumerate(
|
167 |
+
zip(image_paths, label_paths)
|
168 |
+
):
|
169 |
+
image_id = idx
|
170 |
+
annotations = []
|
171 |
+
image = Image.open(image_path) # .convert("RGB")
|
172 |
+
w, h = image.size
|
173 |
+
with open(label_path, "r") as f:
|
174 |
+
lines = f.readlines()
|
175 |
+
for line in lines:
|
176 |
+
line = line.strip().split()
|
177 |
+
# logger.warn(line)
|
178 |
+
category_id = line[
|
179 |
+
0
|
180 |
+
] # int(line[0]) + 1 # you start with annotation id with '1'
|
181 |
+
x_center = float(line[1])
|
182 |
+
y_center = float(line[2])
|
183 |
+
width = float(line[3])
|
184 |
+
height = float(line[4])
|
185 |
+
|
186 |
+
float_x_center = w * x_center
|
187 |
+
float_y_center = h * y_center
|
188 |
+
float_width = w * width
|
189 |
+
float_height = h * height
|
190 |
+
|
191 |
+
min_x = int(float_x_center - float_width / 2)
|
192 |
+
min_y = int(float_y_center - float_height / 2)
|
193 |
+
width = int(float_width)
|
194 |
+
height = int(float_height)
|
195 |
+
|
196 |
+
annotation = create_annotation_from_yolo_format(
|
197 |
+
min_x,
|
198 |
+
min_y,
|
199 |
+
width,
|
200 |
+
height,
|
201 |
+
image_id,
|
202 |
+
category_id,
|
203 |
+
image_id,
|
204 |
+
# segmentation=opt.box2seg,
|
205 |
+
)
|
206 |
+
annotations.append(annotation)
|
207 |
+
# annotation_id += 1
|
208 |
+
|
209 |
+
# image_id += 1 # if you finished annotation work, updates the image id.
|
210 |
+
example = {
|
211 |
+
"image_id": image_id,
|
212 |
+
"image": image,
|
213 |
+
"width": w,
|
214 |
+
"height": h,
|
215 |
+
"objects": annotations,
|
216 |
+
}
|
217 |
+
yield idx, example
|
218 |
+
if self.config.name == "YOLO":
|
219 |
+
for idx, (image_path, label_path) in enumerate(
|
220 |
+
zip(image_paths, label_paths)
|
221 |
+
):
|
222 |
+
im = Image.open(image_path)
|
223 |
+
width, height = im.size
|
224 |
+
image_id = idx
|
225 |
+
annotations = []
|
226 |
+
with open(label_path, "r") as f:
|
227 |
+
lines = f.readlines()
|
228 |
+
objects = []
|
229 |
+
for line in lines:
|
230 |
+
line = line.strip().split()
|
231 |
+
bbox_class = int(line[0])
|
232 |
+
bbox_xcenter = int(float(line[1]) * width)
|
233 |
+
bbox_ycenter = int(float(line[2]) * height)
|
234 |
+
bbox_width = int(float(line[3]) * width)
|
235 |
+
bbox_height = int(float(line[4]) * height)
|
236 |
+
objects.append(
|
237 |
+
{
|
238 |
+
"label": bbox_class,
|
239 |
+
"bbox": [
|
240 |
+
bbox_xcenter,
|
241 |
+
bbox_ycenter,
|
242 |
+
bbox_width,
|
243 |
+
bbox_height,
|
244 |
+
],
|
245 |
+
}
|
246 |
+
)
|
247 |
|
248 |
+
yield idx, {
|
249 |
+
"image": image_path,
|
250 |
+
"objects": objects,
|
251 |
+
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|