tqhuyen commited on
Commit
aa46b7b
1 Parent(s): 9491448

Upload my_dataset.py

Browse files
Files changed (1) hide show
  1. my_dataset.py +37 -0
my_dataset.py CHANGED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import json
2
+
3
+ import datasets
4
+
5
+
6
+ datasets.Features(
7
+ {
8
+ "img_id": datasets.Image("path"),
9
+ "anno_polygons": datasets.Sequence(
10
+ {
11
+ [
12
+ {'category_id': datasets.Value("int32"),
13
+ 'segmentation': datasets.Sequence(feature=list, ),
14
+ 'area': datasets.Value("int64"),
15
+ 'bbox': datasets.Sequence('list'),
16
+ 'width': datasets.Value('int32'),
17
+ 'height': datasets.Value('int32')
18
+ },
19
+ ]
20
+ }
21
+ ),
22
+ "anno_texts": datasets.Value("string"),
23
+ "anno_labels": datasets.Value("string"),
24
+ "anno_num": datasets.Value("int32"),
25
+ "anno_image_quality": datasets.Value("float64")
26
+ }
27
+ )
28
+
29
+
30
+ _URL = ""
31
+ def _split_generators(self, dl_manager):
32
+
33
+ return [
34
+ datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": ".\mcocr_public_train_test_shared_data\mcocr_train_data"}),
35
+ datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": ".\mcocr_public_train_test_shared_data\mcocr_val_data"}),
36
+ ]
37
+