import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import cv2
#import keras
import gradio as gr
import keras
from tensorflow.keras.models import load_model 

# Modellerin eklenmesi

SHAPE = (224, 224, 3)

predictor_disease_risk = load_model('predictor_Disease_Risk.h5')
predictor_dr = load_model('predictor_DR.h5')
predictor_dn = load_model('predictor_DN.h5')

# Fonksiyonlar
# Veri düzenleme, normalizasyon

def cut_and_resize(image):
  LOW_TOL = 20
  img_bw = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
  img_bw[img_bw<=LOW_TOL] = 0
  y_nonzero, x_nonzero = np.nonzero(img_bw)
  image = image[np.min(y_nonzero):np.max(y_nonzero), np.min(x_nonzero): np.max(x_nonzero), ]
  return cv2.resize(image, SHAPE[:2], interpolation = cv2.INTER_LINEAR)
 
def simple_normalizer(X):
  return X / 255.0

# Tanı - Veri okuma ve Verilerin DL Modellerden geçirilmesi

def predict (image_path):
  image = simple_normalizer(cut_and_resize(cv2.imread(image_path)))
  result = predictor_disease_risk.predict(np.array([image]))[0][0] 
  
  dr = predictor_dr.predict(np.array([image]))[0][0]  
  dn = predictor_dn.predict(np.array([image]))[0][0]

  if dr>0.05 and dn<dr:
      result == dr
  elif dn>0.05 and dr<dn:
      result == dn
  else:
      result == 0
          
  diseases = {'Diyabet':float(result)}
  
  to_delete = []
  for k,v in diseases.items():
    if v < 0.05:
      to_delete.append(k)
  
  for k in to_delete:
    del diseases[k]
  
  if len(diseases) == 0:
    diseases = {'Hastalık bulunamadı': 0.0}
  
  return (
    {'Hasta': float(result), 'Sağlıklı': 1 - float(result)}, diseases
  )

title = 'Otomatik Tanı Sistemi'
description = 'Retina görüntüsünden Diyabet ve Hipertansiyon teşhisi yapar.'

interface = gr.Interface(
  predict,
  inputs = [gr.inputs.Image(source="upload",type="filepath", label="Imagen")],
  outputs= [gr.outputs.Label(num_top_classes=2, label='Retina'), gr.outputs.Label(num_top_classes=4, label='Hastalık')],
  title = title, description = description,
  theme = 'peach',
  examples = ['10.png', '15.png', 'retina_normal.jpeg']
)

interface.launch()