Update app.py

app.py

@@ -2,30 +2,28 @@ import gradio as gr
 import tensorflow as tf
 import numpy as np
 from PIL import Image
-from huggingface_hub import hf_hub_download
 
-# Download the model from HF Files
-model_path = hf_hub_download(repo_id="username/space_name", filename="brain_tumor_model.h5")
-model = tf.keras.models.load_model(model_path)
+# Load the model from local path (available in Space's files)
+model = tf.keras.models.load_model("brain_tumor_model.h5")
 
 # Class labels
 class_names = ['Glioma Tumor', 'Meningioma Tumor', 'No Tumor', 'Pituitary Tumor']
 
-# Preprocessing function
+# Image preprocessing
 def preprocess_image(image):
     image = image.resize((224, 224))
     image = np.array(image) / 255.0
     if image.shape[-1] == 4:
-        image = image[..., :3]
+        image = image[..., :3]  # Remove alpha if present
     image = np.expand_dims(image, axis=0)
     return image
 
-# Prediction function
+# Prediction logic
 def predict(image):
     img = preprocess_image(image)
-    predictions = model.predict(img)
-    predicted_class = class_names[np.argmax(predictions)]
-    confidence = np.max(predictions)
+    prediction = model.predict(img)
+    predicted_class = class_names[np.argmax(prediction)]
+    confidence = np.max(prediction)
     return f"Predicted Tumor Type: {predicted_class} (Confidence: {confidence:.2f})"
 
 # Gradio interface
@@ -33,8 +31,8 @@ interface = gr.Interface(
     fn=predict,
     inputs=gr.Image(type="pil"),
     outputs="text",
-    title="Brain Tumor Classifier",
-    description="Upload a Brain MRI image. The model will classify it as Glioma, Meningioma, Pituitary Tumor, or No Tumor."
+    title="🧠 Brain Tumor MRI Classifier",
+    description="Upload a Brain MRI image to predict if it is Glioma, Meningioma, Pituitary Tumor, or No Tumor."
 )
 
 interface.launch()