Update app.py

app.py

@@ -1,32 +1,40 @@
-import os
-os.environ["PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION"] = "python"
 import gradio as gr
 import tensorflow as tf
 import numpy as np
 from PIL import Image
 
+#!pip install tensorflow tensorflow-datasets gradio pillow matplotlib
 
-# Modell laden
-model = tf.keras.models.load_model('pokemon_classifier_model.h5')
+model_path = "pokemon-model_transferlearning.keras"
+model = tf.keras.models.load_model(model_path)
 
-class_names = ['Chansey', 'Growlithe', 'Lapras']
+# Define the core prediction function
+def predict_pokemon(image):
+    # Preprocess image
+    image = Image.fromarray(image.astype('uint8'))  # Convert numpy array to PIL image
+    image = image.resize((150, 150)) # Resize the image to 150x150
+    image = np.array(image)
+    image = np.expand_dims(image, axis=0) # Add batch dimension
+    
+    # Predict
+    prediction = model.predict(image)
+    
+    # Apply softmax to get probabilities for each class
+    probabilities = tf.nn.softmax(prediction)
+    
+    # Map probabilities to Pokemon classes
+    pokemon_classes = ['Articuno', 'Bulbasaur', 'Charmander']  
+    probabilities_dict = {pokemon_class: round(float(probability), 2) for pokemon_class, probability in zip(pokemon_classes, probabilities[0])}
+    
+    return probabilities_dict
 
-def predict(image):
-    image = image.resize((150, 150))
-    img_array = tf.keras.preprocessing.image.img_to_array(image)
-    img_array = np.expand_dims(img_array, axis=0)
-    predictions = model.predict(img_array)
-    score = tf.nn.softmax(predictions[0])
-    return {class_names[i]: float(score[i]) for i in range(3)}
-
-# Gradio Interface
-interface = gr.Interface(
-    fn=predict,
-    inputs=gr.inputs.Image(shape=(150, 150)),
-    outputs=gr.outputs.Label(num_top_classes=3),
-    title="Pokémon Classifier",
-    description="Upload an image of Chansey, Growlithe, or Lapras"
-)
-
-if __name__ == "__main__":
-    interface.launch()
+# Create the Gradio interface
+input_image = gr.Image()
+iface = gr.Interface(
+    fn=predict_pokemon,
+    inputs=input_image, 
+    outputs=gr.Label(),
+    live=True,
+    examples=["images/01.jpg", "images/02.png", "images/03.png", "images/04.jpg", "images/06.png", "images/06.png"],   
+    description="A simple mlp classification model for image classification using the mnist dataset.")
+iface.launch()