app.py

import transformers
from transformers import ViTModel, ViTImageProcessor
from torchvision import transforms
import gradio as gr

model = ViTModel.from_pretrained("afern24/vit-beans-finetuned")

def classify_image(inp):
  inp = transforms.ToTensor()(inp).unsqueeze(0)
  processor = ViTImageProcessor.from_pretrained('google/vit-base-patch16-224-in21k')
  inputs = processor(images=inp, return_tensors="pt")
  outputs = model(**inputs)
  return outputs

# Define your examples as a list of lists
examples = [["leaf1.jpg"], ["leaf2.jpg"], ["leaf3.jpg"]]

# Load your interface with examples
gr.Interface(fn=classify_image, inputs=gr.inputs.Image(type="pil"), outputs=gr.outputs.Label(num_top_classes=3), examples=examples).launch(debug=True)