---
datasets:
- pawlo2013/kimchi-dataset
metrics:
- accuracy
---
Although it's  unlikely you'll ever use it, This model was trained for classifying different types of kimchi (Korean cabbage). It achieved 89% accuracy on a test set comprising of 11 different classes of this Korean delicacy.

labels_eng = ['baechu', 'baik', 'boochoo', 'chongkak', 'got', 'kkakdoogi', 'moosaengchae', 'nabak', 'ohyeesobaki', 'pa', 'yeolmoo'] </br>
labels_kr = ['배추', '백(대파)', '부추', '총각김치', '고추', '깍두기', '무생채', '나박', '오이소박이', '파', '열무']

This model is basically a fintuned version of <a href ='https://huggingface.co/google/vit-base-patch16-224-in21k'> the vision tranformer </a>.

<img src="training_labels.png"> </img>