Sample running code
from transformers import VisionEncoderDecoderModel, ViTImageProcessor, GPT2Tokenizer
import torch
from PIL import Image
model = VisionEncoderDecoderModel.from_pretrained("evlinzxxx/my_model_ViTB-16")
feature_extractor = ViTImageProcessor.from_pretrained("evlinzxxx/my_model_ViTB-16")
tokenizer = GPT2Tokenizer.from_pretrained("evlinzxxx/my_model_ViTB-16")
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)
def show_image_and_captions(url):
display(load_image(url))
our_caption = get_caption(model, image_processor, tokenizer, url)
print(f"Our caption: {our_caption}")
show_image_and_captions("/content/drive/MyDrive/try/test_400/gl_16.jpg")