import cv2
import numpy as np
import streamlit as st
import tensorflow as tf
from PIL import Image
from tensorflow.keras.preprocessing import image
from keras.preprocessing.image import img_to_array
from tensorflow.keras.applications.mobilenet_v2 import MobileNetV2,preprocess_input as mobilenet_v2_preprocess_input

cnn = tf.keras.models.load_model("saved_model/cnn_brain_model_9_11.h5")
resnet = tf.keras.models.load_model("saved_model/resnet_brain_model_9_11.h5")
vgg = tf.keras.models.load_model("saved_model/vgg_brain_model_9_11.h5")

### load file
uploaded_file = st.file_uploader("Choose a image file", type="jpg")

map_dict = {0: 'demented',
            1: 'glioma',
            2: 'healthy',
            3: 'meningioma',
            4: 'pituitary'}
if uploaded_file is not None:
    
    file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
    opencv_image = cv2.imdecode(file_bytes, 1)
    image = Image.fromarray(opencv_image, 'RGB')
    
    image = image.resize((200, 200))
    image=np.array(image)
    input_img = np.expand_dims(image, axis=0)
    st.image(image, channels="RGB")
  
    img=[]
    img.append(input_img)
    img = np.array(img)
   
    print(img.shape)
    
    img = img.astype(np.float32) / 255.0 
    
    Genrate_pred = st.button("Generate Prediction")    
    if Genrate_pred:
        prediction = cnn.predict(img[0]).argmax()
        st.title("Predicted Label for the image by cnn model is {}".format(map_dict [prediction]))
        prediction = resnet.predict(img[0]).argmax()
        st.title("Predicted Label for the image by resnet model is {}".format(map_dict [prediction]))
        prediction = vgg.predict(img[0]).argmax()
        st.title("Predicted Label for the image by vgg model is {}".format(map_dict [prediction]))