### 1. Imports and class names setup ###

# Imports
import gradio as gr
import os
import torch

from pathlib import Path
from model import create_effnetb2_model
from timeit import default_timer as timer
from typing import Tuple, Dict

# Setup class names
class_names = ['pizza', 'steak', 'sushi']

### 2. Model and Transforms Preparation ###
effnetb2, effnetb2_transforms = create_effnetb2_model(num_classes = 3)

# Load save weights
BASE_DIR = Path(__file__).resolve().parent
WEIGHTS_PATH = BASE_DIR / "09_pretrained_effnetb2_pizza_steak_sush_20_percent.pth"

effnetb2.load_state_dict(
    torch.load(WEIGHTS_PATH, map_location=torch.device("cpu"))
)

### 3. Predict function ###
def predict(img) -> Tuple[Dict, float]:
  # Timer
  start_time = timer()

  # Transform the input image to work with EffnetB2
  img = effnetb2_transforms(img).unsqueeze(0) # unsqueeze = add batch dimension on 0th index

  # Eval mode and torch inference mode on
  effnetb2.eval()
  with torch.inference_mode():
    # Pass transfomed image through the moedl and turn prediction logits into probbabilitites
    pred_probs = torch.softmax(effnetb2(img), dim = 1)

  # Create preiciton label and predicition probability dictionary
  pred_labels_and_probs = {class_names[i]: float(pred_probs[0][i]) for i in range(len(class_names))}

  # Calculate prediction time
  end_time = timer()
  pred_time = round(timer() - start_time, 3)

  # Return pred dict and pred time
  return pred_labels_and_probs, pred_time

### 4. Gradio app ###

# Create title for the gradio
title = "FoodVision Mini - Erdem Atak Version"
description = "EfficientNetB2 computer vision model to classify food images"
article = "PyTorch Model Deployment"

# Create example list
example_list = [["examples/" + example] for example in os.listdir("examples")]

# Create the gradio demo
demo = gr.Interface(fn = predict, # it maps inputs to outputs
                    inputs = gr.Image(type = "pil"),
                    outputs = [gr.Label(num_top_classes = 3,
                                       label = "Predictions"),
                              gr.Number(label = "Prediction Time (s)")],
                    examples = example_list, # exaple list above
                    title = title,
                    description = description,
                    article = article)

# launch the demo
demo.launch(debug = False,
            share = True ) # public shareable URL