Datasets:

Languages:
English
Size Categories:
10K<n<100K
Annotations Creators:
machine-generated
Source Datasets:
extended|mnist
ArXiv:
Tags:
License:
mweiss commited on
Commit
705d9cf
1 Parent(s): c87b5df

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -33,10 +33,10 @@ We provide four splits:
33
 
34
  - `test`: 10'000 ambiguous images
35
  - `train`: 10'000 ambiguous images - adding ambiguous images to the training set makes sure test-time ambiguous images are in-distribution.
36
- - `test_mixed`: 20'000 images, consisting of the (shuffled) concatenation of our ambiguous `test` test and the nominal mnist test set by LeCun et. al.,
37
  - `train_mixed`: 70'000 images, consisting of the (shuffled) concatenation of our ambiguous `training` and the nominal training set.
38
 
39
- Note that the ambiguous train images are highly ambiguous (i.e., the two classes have very similar ground truth likelihoods),
40
  the training set images allow for more unbalanced ambiguity.
41
  This is to make the training set more closely connected to the nominal data, while still keeping the test set clearly ambiguous.
42
 
 
33
 
34
  - `test`: 10'000 ambiguous images
35
  - `train`: 10'000 ambiguous images - adding ambiguous images to the training set makes sure test-time ambiguous images are in-distribution.
36
+ - `test_mixed`: 20'000 images, consisting of the (shuffled) concatenation of our ambiguous `test` set and the nominal mnist test set by LeCun et. al.,
37
  - `train_mixed`: 70'000 images, consisting of the (shuffled) concatenation of our ambiguous `training` and the nominal training set.
38
 
39
+ Note that the ambiguous test images are highly ambiguous (i.e., the two classes have very similar ground truth likelihoods),
40
  the training set images allow for more unbalanced ambiguity.
41
  This is to make the training set more closely connected to the nominal data, while still keeping the test set clearly ambiguous.
42