app.py

import streamlit as st
from tensorflow.keras.models import load_model
from predictor import predict_cat_dog
import tempfile
from PIL import Image

# ---------------------------
# Page configuration
# ---------------------------
st.set_page_config(
    page_title="Cat vs Dog Image Classifier",
    page_icon="🐱",
    layout="centered"
)

# ---------------------------
# Title & description
# ---------------------------
st.title("🐱🐶 Cat vs Dog Image Classification")
st.markdown(
    """
    This project uses a **Convolutional Neural Network (CNN)** trained on image data  
    to classify whether an uploaded image contains a **Cat** or a **Dog**.
    """
)

# ---------------------------
# Load model once (important)
# ---------------------------
@st.cache_resource
def load_cnn_model():
    return load_model("cat_vs_dog")

model = load_cnn_model()

# ---------------------------
# Image uploader
# ---------------------------
uploaded_file = st.file_uploader(
    "Upload an image (JPG / PNG)",
    type=["jpg", "jpeg", "png"]
)

# ---------------------------
# Prediction logic
# ---------------------------
if uploaded_file is not None:
    image = Image.open(uploaded_file)
    st.image(image, caption="Uploaded Image", use_column_width=True)

    # Save image temporarily for prediction
    with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as tmp:
        image.save(tmp.name)
        img_path = tmp.name

    label, confidence = predict_cat_dog(img_path, model)

    # ---------------------------
    # Display result
    # ---------------------------
    st.markdown("---")
    st.subheader("🧠 Model Prediction")

    st.markdown(f"### **{label}**")
    st.progress(confidence)

    st.markdown(f"**Confidence:** {confidence:.2f}")

# ---------------------------
# Footer
# ---------------------------
st.markdown("---")
st.markdown(
    "<center><small>Built with TensorFlow & Streamlit</small></center>",
    unsafe_allow_html=True
)