metadata
license: apache-2.0
language:
- en
metrics:
- bleu
pipeline_tag: image-to-text
About model
The model fined tuning with large data of api bbcgoodfood.com
How to Get Started with the Model
from transformers import AutoTokenizer, VisionEncoderDecoderModel, ViTFeatureExtractor
tokenizer = AutoTokenizer.from_pretrained("gpt2") # for text
feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224") # for image
model = VisionEncoderDecoderModel.from_pretrained("sunbv56/vit-gpt2-imagecaptioningfood") # load model
Example code here
https://www.kaggle.com/code/thuntrngbnh/test-model-vit-gpt2-icf/notebook