app.py

import streamlit as st
import numpy as np
import joblib
import pickle
from PIL import Image
from tensorflow.keras.preprocessing.image import img_to_array
from tensorflow.keras.applications.mobilenet_v2 import preprocess_input, MobileNetV2
from tensorflow.keras.models import Model

# Constants
IMAGE_SIZE = (128, 128)

# Load the trained KNN model
knn_model = joblib.load("knn_animal_classifier.pkl")

# Load class labels
with open("class_labels.pkl", "rb") as f:
    class_labels = pickle.load(f)

# Load the MobileNetV2 feature extractor
base_model = MobileNetV2(weights="imagenet", include_top=False, input_shape=(128, 128, 3), pooling="avg")
feature_extractor = Model(inputs=base_model.input, outputs=base_model.output)

# Streamlit UI
st.title("🐾 Animal Image Classifier (KNN + MobileNetV2)")
st.write("Upload an image of an animal to classify it.")

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

if uploaded_file is not None:
    image = Image.open(uploaded_file).convert("RGB")
    st.image(image, caption="Uploaded Image", use_column_width=True)

    # Preprocess image
    img = image.resize(IMAGE_SIZE)
    img_array = img_to_array(img)
    img_array = preprocess_input(img_array)
    img_array = np.expand_dims(img_array, axis=0)

    # Extract features and predict
    features = feature_extractor.predict(img_array)
    prediction = knn_model.predict(features)[0]

    st.success(f"🧠 Predicted Animal: **{prediction}**")