app.py

import streamlit as st
import tensorflow as tf
from PIL import Image
import os

model = tf.keras.models.load_model('Brain_tumor/')
st.write('Model is loaded successfully')

TEMP_DIR = 'temp'
if not os.path.exists(TEMP_DIR):
    os.makedirs(TEMP_DIR)

class_names = ['glioma', 'meningioma', 'notumor', 'pituitary']

def load_and_prep_imgg(filename, img_shape=229, scale=True):
    img = tf.io.read_file(filename)
    img = tf.io.decode_image(img)
    img = tf.image.resize(img, size=[img_shape, img_shape])
    if scale:
        return img / 255
    else:
        return img

st.title('Brain Tumor Classification Prediction using Xception ImageNet')

uploaded_file = st.sidebar.file_uploader('Upload your Image', type=['jpg'])

if uploaded_file:
    file_path = os.path.join(TEMP_DIR, uploaded_file.name)
    
    # Save the uploaded file to the temporary directory
    with open(file_path, "wb") as f:
        f.write(uploaded_file.getbuffer())
    
    img = load_and_prep_imgg(file_path, scale=True)
    imgg = Image.open(file_path)
    st.image(imgg, caption="Uploaded Image")

    pred_img = model.predict(tf.expand_dims(img, axis=0))
    pred_class = class_names[pred_img.argmax()]
    st.write(f"Predicted brain tumor is: {pred_class} with probability: {pred_img.max():.2f}")