zrthxn/diffusion
Updated
image
imagewidth (px) 64
64
| label
class label 3
classes |
|---|---|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
|
0non_smile
|
End of preview. Expand
in Dataset Viewer.
YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/datasets-cards)
Similing or Not
A dataset comprised of closeups of people's faces, belonging to 2 binary classes.
- 600 smiling faces in the "smile" folder.
- 603 non smiling faces in the "non_smile" folder.
We can build a smile detector with this dataset, and even a "smile transformer" via a Style Transfer algorithm.
The "test" folder contains ~12k unlabeled faces. If someone wants go through the work of labeling these faces as smile/nonsmile and republish a greater version of this dataset, please be my guest!
Reupload from original dataset on Kaggle
- Downloads last month
- 2
Size of downloaded dataset files:
23.8 MB
Size of the auto-converted Parquet files:
11.7 MB
Number of rows:
13,233
Models trained or fine-tuned on zrthxn/SmilingOrNot