datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
top_quark_taggingTop Quark Tagging is a dataset of Monte Carlo simulated hadronic top and QCD dijet events for the evaluation of top quark tagging architectures. The dataset consists of 1.2M training events, 400k validation events and 400k test events.X-Ray_Community_Tagging
What is this?
A community effort is a must in order to make a better, more accurate vision model, as I simply cannot tag thousands of images. If you would provide 50 corrections and 20 more people do so as well, it would help a lot.
If 100 ppl would help with 50 corrections each, we might have a high-accuracy functioning uncensored vision model.
The best format would be to name the output and images with the same name, like:
1.png
1.txt
2.png
2.txt
The best approach is probably… See the full description on the dataset page: https://huggingface.co/datasets/SicariusSicariiStuff/X-Ray_Community_Tagging.top_tagging_imageshotel-tagging-finetune
Hotel Tagging Finetune
Hotel-room images for a tagging finetune workflow where an LLM is used as the judge.
This public release intentionally contains only the images.
Splits
Split
Samples
train
2000
test
250
Columns
image: hotel-room image
240903-image-taggingJarpybooru-taggingsoybooru-tagging