Over 5,000 images of 20 different classes.
This dataset can be freely used for any purpose, including commercial:
For example:
- Creating a tutorial or a course (free or paid)
- Writing a book
- Kaggle competitions (as an external dataset)
- Training an internal model at any company
The images.csv
file contains:
image
- the ID of the image (use it to load the image fromimages/<ID>.jpg
)sender_id
- the ID of a person who contributed the imagelabel
- the class of the imagekids
- flag,True
if it's clothes for kids
- If you're looking for a subset of the clothing dataset, check here: https://github.com/alexeygrigorev/clothing-dataset-small
- You can read more about this dataset here: https://medium.com/data-science-insider/clothing-dataset-5b72cd7c3f1f
- This dataset is also awailable on Kaggle (with images in higher resolution): https://www.kaggle.com/agrigorev/clothing-dataset-full/ (please upvote it!)
Images of some classes don’t appear very often. Training a neural network to predict these classes is quite difficult — we need at least 100-200 images of each class to make a meaningful model.
That’s why, for educational purposes, we created a subset of the full dataset that covers only the top-10 classes.
Check it here: https://github.com/alexeygrigorev/clothing-dataset-small
- https://www.kaggle.com/agrigorev/collage
- Train a self-supervised model using lightly on the clothing-dataset
- Add a link here
Do you use this dataset somewhere? Please submit a PR with a link
We'd like to thank
- Kenes Shangereyev and Tagias.com for helping with 3000 images
- All the 32 people who contributed their images to the dataset via the forms:
- Everyone who supported the initiative by engaging with the announcements on social media
It wouldn't be possible to collect this dataset without your help!