tensorflow / tensorflow/datasets

emnist/letters dataset has 26 classes not 37

Open
#2,579 5 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug question
Dominant language
Python
Stars
4.6k
Forks
1.6k
Avg merge
3h 54m
Merged PRs (30d)
1

Description

Description of issue
emnist/letters says it has 37 classes but it really only has 26

Describe what the problem is. Please include URL(s) if possible.

https://www.tensorflow.org/datasets/catalog/emnist
On the above URL the the emnist/letters dataset is described as having 37 classes. It actually only has 26 classes. All upper and lower case letters were merged together in the emnist/letters dataset.

This is not simply a typo issue because the meta data also states that there are 37 classes when it should report only 26. So I guess there are 37 classes but some classes are never used.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the EMNIST catalog entry linked in the issue and inspect how the letters dataset metadata declares its class count. Compare that declaration with the labels actually present, then update the affected metadata so the catalog and dataset report the correct count.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
data, machine-learning
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.