Skip to content

Fixes EMNIST split and label issues #2673

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Oct 23, 2020
Merged

Fixes EMNIST split and label issues #2673

Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
126c594
Add float support to ColorJitter
vballoli Sep 13, 2020
aabc455
Fix byclass EMNIST
vballoli Sep 13, 2020
945e056
Fix bymerge, balance, letters EMNIST
vballoli Sep 13, 2020
87b2548
Fix whitespace indent
vballoli Sep 13, 2020
4d8762b
Revert unrelated file changes
vballoli Sep 13, 2020
8bf3c05
Revert unrelated file changes
vballoli Sep 13, 2020
8cb4531
Removing unnecessary type conversions.
datumbox Oct 23, 2020
fa7dd62
Removing the transform and adding dummy class instead.
datumbox Oct 23, 2020
9c6a2d8
Merge branch 'master' into bugfix/emnist
datumbox Oct 23, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions torchvision/datasets/mnist.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -249,13 +249,13 @@ class EMNIST(MNIST):
md5 = "58c8d27c78d21e728a6bc7b3cc06412e"
splits = ('byclass', 'bymerge', 'balanced', 'letters', 'digits', 'mnist')
# Merged Classes assumes Same structure for both uppercase and lowercase version
_merged_classes = set(['C', 'I', 'J', 'K', 'L', 'M', 'O', 'P', 'S', 'U', 'V', 'W', 'X', 'Y', 'Z'])
_all_classes = set(list(string.digits + string.ascii_letters))
_merged_classes = {'c', 'i', 'j', 'k', 'l', 'm', 'o', 'p', 's', 'u', 'v', 'w', 'x', 'y', 'z'}
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

One remark concerning cosmetics, not actual functionality. The EMNIST paper describes removing the capital letters. We can actually write the classes in a way that achieves this but then the labels will look weird; something like: A, B, c, D, .... So I think @vballoli's solution is fine.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, I think this is fine as is, but thanks for pointing it out

_all_classes = set(string.digits + string.ascii_letters)
classes_split_dict = {
'byclass': list(_all_classes),
'byclass': sorted(list(_all_classes)),
'bymerge': sorted(list(_all_classes - _merged_classes)),
'balanced': sorted(list(_all_classes - _merged_classes)),
'letters': list(string.ascii_lowercase),
'letters': ['N/A'] + list(string.ascii_lowercase),
'digits': list(string.digits),
'mnist': list(string.digits),
}
Expand Down