Datasets:

Modalities:
Image
Formats:
parquet
Languages:
English
DOI:
Libraries:
Datasets
Dask
License:
jpodivin commited on
Commit
47645d8
1 Parent(s): 630efe3

Adding loading script

Browse files

Signed-off-by: Jiri Podivin <[email protected]>

Files changed (1) hide show
  1. plantorgans.py +85 -0
plantorgans.py ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import datasets
2
+
3
+ _DESCRIPTION = """Photos of various plants with their major, above ground organs labeled. Includes labels for stem, leafs, fruits and flowers."""
4
+
5
+ _HOMEPAGE = "https://huggingface.co/datasets/jpodivin/plantorgans"
6
+
7
+ _CITATION = """"""
8
+
9
+ _LICENSE = "MIT"
10
+
11
+ _NAMES = [
12
+ 'Leaf',
13
+ 'Stem',
14
+ 'Flower',
15
+ 'Fruit',
16
+ ]
17
+
18
+ _BASE_URL = "https://huggingface.co/datasets/jpodivin/plantorgans/blob/main/sourcedata/"
19
+
20
+ _DOWNLOAD_SUFFIX = "download?true"
21
+ _METADATA_URLS = [
22
+ ""
23
+ ]
24
+
25
+ class PlantOrgans(datasets.GeneratorBasedBuilder):
26
+ """Plantorgans dataset
27
+ """
28
+ BUILDER_CONFIGS = [
29
+ PlantOrgansConfig(
30
+ name="semantic_segmentation",
31
+ description="This configuration contains segmentation masks."
32
+ ),
33
+ ]
34
+
35
+ def _info(self):
36
+ return datasets.DatasetInfo(
37
+ description=_DESCRIPTION,
38
+ features=datasets.Features(
39
+ {
40
+ "image": datasets.Image(),
41
+ "annotation": datasets.ClassLabel(names=_NAMES),
42
+ }
43
+ ),
44
+ supervised_keys=("image", "annotation"),
45
+ homepage=_HOMEPAGE,
46
+ citation=_CITATION,
47
+ license=_LICENSE,
48
+ )
49
+
50
+
51
+ def _split_generators(self, dl_manager):
52
+ archive_path = dl_manager.download(_BASE_URL)
53
+ split_metadata_paths = dl_manager.download(_METADATA_URLS)
54
+ return [
55
+ datasets.SplitGenerator(
56
+ name=datasets.Split.TRAIN,
57
+ gen_kwargs={
58
+ "images": dl_manager.iter_archive(archive_path),
59
+ "metadata_path": split_metadata_paths["train"],
60
+ },
61
+ ),
62
+ datasets.SplitGenerator(
63
+ name=datasets.Split.VALIDATION,
64
+ gen_kwargs={
65
+ "images": dl_manager.iter_archive(archive_path),
66
+ "metadata_path": split_metadata_paths["test"],
67
+ },
68
+ ),
69
+ ]
70
+ def _generate_examples(self, images, metadata_path):
71
+ ...
72
+
73
+ class PlantOrgansConfig(datasets.BuilderConfig):
74
+ """Builder Config for PlantOrgans"""
75
+
76
+ def __init__(self, data_url, metadata_urls, **kwargs):
77
+ """BuilderConfig for PlantOrgans.
78
+ Args:
79
+ data_url: `string`, url to download the zip file from.
80
+ metadata_urls: dictionary with keys 'train' and 'validation' containing the archive metadata URLs
81
+ **kwargs: keyword arguments forwarded to super.
82
+ """
83
+ super().__init__(version=datasets.Version("1.0.0"), **kwargs)
84
+ self.data_url = data_url
85
+ self.metadata_urls = metadata_urls