Datasets:
Tasks:
Image Classification
Modalities:
Image
Formats:
parquet
Languages:
English
Size:
1K - 10K
Tags:
veterinary
cell classification
animal
endangered species
microscopy
microscopic biological image
License:
Create gen_script.py
Browse files- gen_script.py +85 -0
gen_script.py
ADDED
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from pathlib import Path
|
2 |
+
|
3 |
+
import datasets
|
4 |
+
|
5 |
+
_VERSION = "0.1.0"
|
6 |
+
|
7 |
+
_CITATION = ""
|
8 |
+
|
9 |
+
_DESCRIPTION = ""
|
10 |
+
|
11 |
+
_HOMEPAGE = ""
|
12 |
+
|
13 |
+
_LICENSE = ""
|
14 |
+
|
15 |
+
_FEATURES = datasets.Features(
|
16 |
+
{
|
17 |
+
"image": datasets.Image(mode="RGB"),
|
18 |
+
"label": datasets.ClassLabel(
|
19 |
+
names=[
|
20 |
+
"Basophil",
|
21 |
+
"Eosinophil",
|
22 |
+
"Lymphocyte",
|
23 |
+
"Monocyte",
|
24 |
+
"Neutrophil",
|
25 |
+
]
|
26 |
+
),
|
27 |
+
}
|
28 |
+
)
|
29 |
+
|
30 |
+
Cropped = datasets.Split("cropped")
|
31 |
+
Augmented = datasets.Split("augmented")
|
32 |
+
Original = datasets.Split("original")
|
33 |
+
|
34 |
+
|
35 |
+
class WartyPig(datasets.GeneratorBasedBuilder):
|
36 |
+
DEFAULT_WRITER_BATCH_SIZE = 1000
|
37 |
+
|
38 |
+
def _info(self):
|
39 |
+
return datasets.DatasetInfo(
|
40 |
+
features=_FEATURES,
|
41 |
+
supervised_keys=None,
|
42 |
+
description=_DESCRIPTION,
|
43 |
+
homepage=_HOMEPAGE,
|
44 |
+
license=_LICENSE,
|
45 |
+
version=_VERSION,
|
46 |
+
citation=_CITATION,
|
47 |
+
)
|
48 |
+
|
49 |
+
def _split_generators(self, dl_manager: datasets.DownloadManager):
|
50 |
+
original_images = sorted(list(Path("Original").rglob("*.jpg")))
|
51 |
+
augmented_images = sorted(list(Path("Augmented images").rglob("*.jpg")))
|
52 |
+
cropped_images = sorted(list(Path("Cropped Classified").rglob("*.jpg")))
|
53 |
+
|
54 |
+
return [
|
55 |
+
datasets.SplitGenerator(
|
56 |
+
name=Original,
|
57 |
+
gen_kwargs={"images": original_images, "no_label": True},
|
58 |
+
),
|
59 |
+
datasets.SplitGenerator(
|
60 |
+
name=Cropped,
|
61 |
+
gen_kwargs={"images": cropped_images},
|
62 |
+
),
|
63 |
+
datasets.SplitGenerator(
|
64 |
+
name=Augmented,
|
65 |
+
gen_kwargs={"images": augmented_images},
|
66 |
+
),
|
67 |
+
]
|
68 |
+
|
69 |
+
def _generate_examples(self, images: list[Path], no_label=False):
|
70 |
+
for i, image in enumerate(images):
|
71 |
+
if no_label:
|
72 |
+
yield (
|
73 |
+
i,
|
74 |
+
{
|
75 |
+
"image": str(image),
|
76 |
+
},
|
77 |
+
)
|
78 |
+
else:
|
79 |
+
yield (
|
80 |
+
i,
|
81 |
+
{
|
82 |
+
"image": str(image),
|
83 |
+
"label": image.parent.name,
|
84 |
+
},
|
85 |
+
)
|