up

app.py

@@ -1,25 +1,32 @@
 import gradio as gr
 from tensorflow.keras.models import load_model
-from tensorflow.keras.preprocessing.image import img_to_array
-from PIL import Image
+from tensorflow.keras.preprocessing.image import img_to_array, load_img
 import numpy as np
+from PIL import Image
 
-# Load your model
-model = load_model('cats_and_dogs_classifier.h5')
+# Load your model (using a relative path)
+model_path = "cats_and_dogs_classifier.h5"
+model = load_model(model_path)
 
-# Define prediction function
 def predict(image):
-    # Convert Gradio PIL Image to numpy array
-    img = np.array(image)
-    # Resize and preprocess image
-    img = np.array(Image.fromarray(img).resize((224, 224)))
-    img = img_to_array(img)
-    img = np.expand_dims(img, axis=0)
-    # Predict class (0 for cat, 1 for dog)
-    prediction = model.predict(img)
-    class_label = 'cat' if prediction[0][0] < 0.5 else 'dog'
+    # Resize image to match model's expected sizing
+    img = image.resize((150, 150))
+    # Convert image to array and expand dimensions to fit model input requirements
+    img_array = img_to_array(img)
+    img_array = np.expand_dims(img_array, axis=0)
+    # Predict using loaded model
+    prediction = model.predict(img_array)
+    # Convert prediction to label
+    class_label = 'Cat' if prediction[0][0] < 0.5 else 'Dog'
     return class_label
 
 # Create a Gradio interface
-interface = gr.Interface(fn=predict, inputs="image", outputs="text")
-interface.launch()
+iface = gr.Interface(fn=predict, 
+                     inputs=gr.inputs.Image(type='pil', label="Upload a photo of a cat or dog"), 
+                     outputs="text",
+                     title="Cat or Dog Classifier",
+                     description="Upload an image of a cat or dog to classify it.")
+
+if __name__ == "__main__":
+    iface.launch()
+