app.py

import streamlit as st
from PIL import Image
import numpy as np
from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing.image import img_to_array, load_img
import keras

# Streamlit app
st.title("Age and Gender Prediction App (InceptionV3) ❤️")

# Display a message before the app starts
with st.spinner("Loading the app..."):
    # Simulate app loading time
    import time
    time.sleep(2)  # Adjust the sleep time as needed

# File uploader
uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])

# Create an empty element for displaying messages
message_placeholder = st.empty()

# Load saved models lazily
@st.cache_resource
def load_models():
    age_model_inception = load_model('age_model_inception.h5')
    gender_model_inception = load_model('gender_model_inception.h5')
    return age_model_inception, gender_model_inception

# Process and predict function for InceptionV3 models
def process_and_predict_inception(file, age_model, gender_model):
    # Load and resize the image to (299, 299)
    img = load_img(file, target_size=(299, 299))
    img_array = img_to_array(img)
    img_array = img_array.reshape((1, 299, 299, 3))
    img_array /= 255.0  # Normalize the image

    # Predict age and gender
    age = age_model.predict(img_array)
    gender = np.round(gender_model.predict(img_array))[0][0]
    gender = 'female' if gender == 1 else 'male'

    return int(age[0]), gender

if uploaded_file is not None:
    # Display the uploaded image
    st.image(uploaded_file, caption="Uploaded Image.", use_column_width=True)
    st.write("")

    # Display "Please wait" message
    with st.spinner("Classifying... Please wait."):
        # Load models
        age_model_inception, gender_model_inception = load_models()

        # Process and predict
        age, gender = process_and_predict_inception(uploaded_file, age_model_inception, gender_model_inception)

    # Clear the "Please wait" spinner
    st.spinner("")
    
    # Stylish presentation using markdown and CSS
    st.markdown(f"**Predicted Age:** {age} years")
    
    # Use HTML and CSS for styling
    if gender == 'female':
        st.markdown('<p style="color:#FF69B4;font-size:20px;">Predicted Gender: Female ❤️</p>', unsafe_allow_html=True)
    else:
        st.markdown('<p style="color:#6495ED;font-size:20px;">Predicted Gender: Male ❤️</p>', unsafe_allow_html=True)