import datasets
from transformers import AutoFeatureExtractor, AutoModelForImageClassification
import gradio as gr
import torch

dataset = datasets.load_dataset('beans','full_size') # This should be the same as the first line of Python code in this Colab notebook

extractor = AutoFeatureExtractor.from_pretrained("saved_model_files")
model = AutoModelForImageClassification.from_pretrained("saved_model_files")

labels = dataset['train'].features['labels'].names

def classify(im):
  features = extractor(im, return_tensors='pt')
  with torch.no_grad():
    logits = model(features["pixel_values"])[-1]
  #logits = torch.nn.functional.softmax(logits, dim=-1)
  probability = torch.nn.functional.softmax(logits, dim=-1)
  probs = probability[0].detach().numpy()
  confidences = {label: float(probs[i]) for i, label in enumerate(labels)}
  print(confidences)
  return confidences


interface = gr.Interface(fn=classify,
             inputs="image",
             outputs=gr.Label(num_top_classes=3),
             title="Bean leaf health classifier",
             description="This is a bean leaf health classifier app created as part of CoRise End to End Vision application project",
             examples=["https://datasets-server.huggingface.co/assets/beans/--/default/validation/3/image/image.jpg",
                       "https://datasets-server.huggingface.co/assets/beans/--/default/test/20/image/image.jpg",
                       "https://datasets-server.huggingface.co/assets/beans/--/default/validation/97/image/image.jpg"])
# FILL HERE

interface.launch()