from typing import Tuple, Dict
from timeit import default_timer as timer 
import PIL
import torch
import os
import gradio as gr
from model import create_effnet

class_names = ["Pizza 🍕", "Steak 🥩", "Sushi 🍣"]

effnetb2, effnet_transforms = create_effnet(2, len(class_names))

effnetb2.load_state_dict(
    torch.load(
        f="effnet_b2_feature_extractor_20_percent_data.pth",
        map_location=torch.device("cpu"),  # load to CPU
    )
)

def predict(img) -> Tuple[Dict, float]:
  start_time = timer()
  img = effnet_transforms(img).unsqueeze(0)
  effnetb2.eval()
  with torch.inference_mode():
    pred_probs =torch.softmax(effnetb2(img), dim=1)
  pred_labels_and_probs = {class_names[i]:float(pred_probs[0][i]) for i in range(len(class_names))}
  end_time = timer()
  return pred_labels_and_probs, round(end_time - start_time, 3)

title = "FoodVision Mini🍕🥩🍣"
desc = "An EffNetB2 feature extractor that classifies images of pizza, steak and sushi (created at the 'Zero To Mastery Learn PyTorch for Deep Learning' course')"
example_list = ["examples/" + example for example in os.listdir("examples")]

demo = gr.Interface(
    fn=predict,
    inputs=gr.Image(type="pil"),
    outputs=[gr.Label(num_top_classes=3, label="Predictions"), gr.Number(label="Prediction time (s)")],
    examples=example_list,
    title=title,
    description=desc
)

demo.launch()