Update app.py

app.py

@@ -1,6 +1,8 @@
+import cv2
 import gradio as gr
 import tensorflow as tf
 import numpy as np
+from PIL import Image
 
 title = "Welcome on your first sketch recognition app!"
 
@@ -13,14 +15,88 @@ head = (
 # Model yükleniyor
 model = tf.keras.models.load_model("number_recognition_model_colab.keras")
 
-def recognize_digit(image):
-    prediction = model.predict(np.reshape(image, (1, 28, 28))).tolist()[0]
-    return {str(i): prediction[i] for i in range(10)}
 
-sketchpad = gr.Sketchpad(shape=(28, 28))
-gr.Interface(fn=recognize_digit, 
-             inputs=sketchpad,
-             outputs="label",
-             title="Handwritten Digits Classifier",
-             description="This app uses lenet5 for handwritten digits classification").launch()
+img_size = 28
 
+labels = ["zero", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine"]
+
+def predict(img):
+    try:
+        # Enhanced image validation and conversion
+        if img is None:
+            raise ValueError("No image provided")
+        
+        # Convert to numpy array if it's a PIL Image
+        if isinstance(img, Image.Image):
+            img = np.array(img)
+        
+        # Handle base64 image strings
+        elif isinstance(img, str):
+            # Check if it's a base64 data URL
+            if img.startswith('data:image'):
+                # Split and decode base64 part
+                img = img.split(',')[1]
+            
+            # Decode base64 to image
+            try:
+                img = Image.open(io.BytesIO(base64.b64decode(img)))
+                img = np.array(img)
+            except Exception as e:
+                print(f"Base64 decoding error: {e}")
+                raise ValueError("Invalid base64 image")
+        
+        # Validate numpy array
+        if not isinstance(img, np.ndarray):
+            raise ValueError("Input could not be converted to a valid image")
+        
+        # Print initial image details for debugging
+        print(f"Initial image type: {type(img)}, shape: {img.shape}")
+        
+        # Handle color channels
+        if img.ndim == 3:
+            if img.shape[-1] == 3:  # Color image
+                img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+            elif img.shape[-1] == 4:  # RGBA image
+                img = cv2.cvtColor(img, cv2.COLOR_RGBA2GRAY)
+        
+        # Ensure single channel
+        if img.ndim == 2:
+            img = np.expand_dims(img, axis=-1)
+        
+        # Resize and normalize
+        img = cv2.resize(img, (img_size, img_size))
+        img = img.astype('float32') / 255.0
+        img = img.reshape(1, img_size, img_size, 1)
+        
+        # Print processed image details
+        print(f"Processed image shape: {img.shape}")
+        
+        # Get predictions from the model
+        preds = model.predict(img)[0]
+        
+        # Print predictions for debugging
+        print("Predictions:", preds)
+        
+        # Return predictions as a dictionary
+        return {label: float(pred) for label, pred in zip(labels, preds)}
+    
+    except Exception as e:
+        # Comprehensive error logging
+        print(f"Full error during prediction: {e}")
+        return {"Error": str(e)}
+            
+   
+
+# Set up the Gradio interface with the input as a sketchpad and output as labels
+label = gr.Label(num_top_classes=3)
+
+# Gradio arayüzü
+interface = gr.Interface(
+    fn=predict,
+    inputs=gr.Sketchpad(type="pil"),
+    outputs=label,
+    title="Sketch Recognition App",
+    description="Draw a number (0-9) and see the model's top predictions."
+)
+
+interface.launch(debug=True, share=True)