zh-plus commited on
Commit
58d3170
·
1 Parent(s): b427d14

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +46 -0
README.md CHANGED
@@ -1,3 +1,49 @@
1
  ---
2
  license: afl-3.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: afl-3.0
3
  ---
4
+
5
+ [Needs More Information]
6
+
7
+ # Dataset Card for tiny-imagenet
8
+
9
+ ## Dataset Description
10
+
11
+ - **Homepage:** https://www.kaggle.com/c/tiny-imagenet
12
+ - **Repository:** [Needs More Information]
13
+ - **Paper:** http://cs231n.stanford.edu/reports/2017/pdfs/930.pdf
14
+ - **Leaderboard:** https://paperswithcode.com/sota/image-classification-on-tiny-imagenet-1
15
+ - **Point of Contact:** [Needs More Information]
16
+
17
+ ### Dataset Summary
18
+
19
+ Tiny ImageNet contains 100000 images of 200 classes (500 for each class) downsized to 64×64 colored images. Each class has 500 training images, 50 validation images, and 50 test images.
20
+
21
+ ### Supported Tasks and Leaderboards
22
+
23
+ [Needs More Information]
24
+
25
+ ### Languages
26
+
27
+ The class labels in the dataset are in English.
28
+
29
+ ## Dataset Structure
30
+
31
+ ### Data Instances
32
+
33
+ ```json
34
+ {
35
+ 'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=64x64 at 0x1A800E8E190,
36
+ 'label': 15
37
+ }
38
+ ```
39
+
40
+ ### Data Fields
41
+
42
+ - image: A PIL.Image.Image object containing the image. Note that when accessing the image column: dataset[0]["image"] the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the "image" column, i.e. dataset[0]["image"] should always be preferred over dataset["image"][0].
43
+ - label: an int classification label. -1 for test set as the labels are missing.
44
+
45
+ ### Data Splits
46
+
47
+ | | Train | Valid |
48
+ | ------------ | ------ | ----- |
49
+ | # of samples | 100000 | 10000 |