Create app.py

app.py

@@ -0,0 +1,74 @@
+import streamlit as st
+import numpy as np
+from tensorflow.keras.models import load_model
+from tensorflow.keras.preprocessing import image
+from PIL import Image
+import matplotlib.pyplot as plt
+import io
+
+# Load model
+@st.cache_resource
+def load_model_safe():
+    model_path = "fish_freshness_model_retrained_final.h5"
+    try:
+        model = load_model(model_path)
+        return model
+    except Exception as e:
+        st.error(f"Error loading model: {e}")
+        return None
+
+model = load_model_safe()
+
+# Define class names and messages
+class_names = ['Fresh', 'Moderately Fresh', 'Spoiled']
+custom_messages = {
+    'Fresh': (
+        "✅ **Fresh Fish Detected**\n"
+        "- Estimated Age: Less than 1 day old\n"
+        "- Bright eyes, red gills, firm flesh\n"
+        "- Safe for raw or cooked consumption."
+    ),
+    'Moderately Fresh': (
+        "⚠️ **Moderately Fresh Fish Detected**\n"
+        "- Estimated Age: 2–3 days\n"
+        "- Slight odor, softening flesh\n"
+        "- Cook thoroughly before eating."
+    ),
+    'Spoiled': (
+        "🚫 **Spoiled or Unsafe Fish Detected**\n"
+        "- Estimated Age: 4+ days\n"
+        "- Dull eyes, strong odor, possible chemical treatment\n"
+        "- Not safe for consumption. Discard immediately."
+    )
+}
+
+# Streamlit UI
+st.title("🐟 Fish Freshness Classifier")
+st.subheader("Upload an image of a fish to analyze its freshness")
+
+uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
+
+if uploaded_file and model:
+    try:
+        img = Image.open(uploaded_file).convert("RGB")
+        img_resized = img.resize((224, 224))  # Match model input size
+        img_array = image.img_to_array(img_resized)
+        img_array = np.expand_dims(img_array / 255.0, axis=0)
+
+        prediction = model.predict(img_array)
+        predicted_index = np.argmax(prediction)
+        confidence = float(np.max(prediction))
+        predicted_class = class_names[predicted_index]
+
+        # Display results
+        st.image(img, caption=f"Uploaded Image", use_column_width=True)
+        st.markdown(f"### 🎯 Prediction: **{predicted_class}** ({confidence*100:.2f}% confidence)")
+        st.markdown(custom_messages[predicted_class])
+
+        # Show confidence for all classes
+        st.subheader("📊 Class Probabilities:")
+        for i, class_name in enumerate(class_names):
+            st.write(f"- {class_name}: {prediction[0][i]*100:.2f}%")
+
+    except Exception as e:
+        st.error(f"Error processing image: {e}")