Update app.py

app.py

@@ -1,80 +1,79 @@
 import os
-# Force TensorFlow to look for the legacy Keras 2 behavior if your model was trained on it
-os.environ["TF_USE_LEGACY_KERAS"] = "1"
+# Force Keras 2 logic for backward compatibility with .h5 models
+os.environ["TF_USE_LEGACY_KERAS"] = "1" 
 
 import gradio as gr
 import tensorflow as tf
-import tf_keras as keras  # Use tf_keras to avoid version mismatch errors
+import tf_keras as keras  # Official "Classic" Keras engine
 import numpy as np
 from PIL import Image
 from huggingface_hub import hf_hub_download
 
 # --- CONFIGURATION ---
-# Replace with your actual Repo ID and .h5 model filename on Hugging Face
-REPO_ID = "mediaportal/BrainTumorDetection" 
-MODEL_FILENAME = "brain99.h5" 
+# IMPORTANT: Change REPO_ID to your actual Hugging Face repository name
+REPO_ID = "mediaportal/BrainTumorMRI" 
+MODEL_FILENAME = "BraintumorMRI99.h5"
 
 model = None
 
 def load_model_with_progress(progress=gr.Progress(track_tqdm=True)):
     global model
     try:
-        progress(0, desc="Downloading model from Hugging Face...")
+        progress(0, desc="Downloading Brain Tumor model...")
         path = hf_hub_download(repo_id=REPO_ID, filename=MODEL_FILENAME)
         
-        progress(0.7, desc="Loading weights using Classic Keras 2 engine...")
-        # Use compile=False to bypass optimizer metadata issues
+        progress(0.7, desc="Loading weights into Xception architecture...")
+        # compile=False avoids errors with custom training optimizers (like Adamax)
         model = keras.models.load_model(path, compile=False)
         
         progress(1.0, desc="✅ Model Ready!")
-        return "Model Loaded Successfully."
+        return "Model Loaded Successfully. Ready to analyze MRI scans."
     except Exception as e:
         return f"❌ Error: {str(e)}"
 
 def predict(img):
     if model is None:
-        return "Please wait for the model to load."
+        return "Please wait for the model to finish loading."
     
     if img is None:
         return "No image provided."
 
-    # 1. Preprocessing: Resize to match the input shape used in your notebook
-    # Your notebook uses 224x224 as standard for these CNN architectures
-    img = Image.fromarray(img.astype('uint8'), 'RGB').resize((224, 224))
+    # 1. Preprocessing: Your notebook used (299, 299) for Xception
+    img = Image.fromarray(img.astype('uint8'), 'RGB').resize((299, 299))
     
-    # 2. Convert to array and normalize (1/255.0)
+    # 2. Rescale: Matches 'rescale=1/255' used in your training ImageDataGenerator
     img_array = np.array(img).astype('float32') / 255.0
     
-    # 3. Add batch dimension (1, 224, 224, 3)
+    # 3. Add batch dimension (1, 299, 299, 3)
     img_array = np.expand_dims(img_array, axis=0)
     
     # Inference
     prediction = model.predict(img_array)[0]
     
-    # Labels based on your notebook's classes
+    # Class labels sorted alphabetically (default behavior of flow_from_dataframe)
     labels = ["Glioma", "Meningioma", "No Tumor", "Pituitary"]
     
-    # Return as a dictionary of label: probability
+    # Return as a dictionary of {Label: Probability}
     return {labels[i]: float(prediction[i]) for i in range(len(labels))}
 
-# --- BUILD GRADIO INTERFACE ---
+# --- GRADIO INTERFACE ---
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
-    gr.Markdown("# 🧠 Brain Tumor Identification from MRI")
-    gr.Markdown("This app uses a Deep Learning model to identify tumor types in MRI scans.")
+    gr.Markdown("# 🧠 Brain Tumor MRI Classification")
+    gr.Markdown("Identify tumor types (Glioma, Meningioma, Pituitary) or healthy scans from MRI images.")
     
     status_box = gr.Markdown("⏳ Initializing system...")
     
     with gr.Row():
         with gr.Column():
             input_img = gr.Image(label="Upload MRI Scan")
-            btn = gr.Button("Analyze MRI", variant="primary")
+            btn = gr.Button("Run Analysis", variant="primary")
         with gr.Column():
-            output_label = gr.Label(num_top_classes=4, label="Diagnosis Prediction")
+            output_label = gr.Label(num_top_classes=4, label="Diagnosis")
 
-    # Automatically load model when app starts
+    # Automatically trigger model download on page load
     demo.load(load_model_with_progress, outputs=status_box)
     
-    # Trigger prediction on button click
+    # Link button to prediction function
     btn.click(fn=predict, inputs=input_img, outputs=output_label)
 
 if __name__ == "__main__":