Update app.py

app.py

@@ -1,31 +1,38 @@
+
 import gradio as gr
-import numpy as np
-from keras.models import load_model
+import tensorflow as tf
 from PIL import Image
+import numpy as np
 
-model = load_model('pokemon-model_2_transferlearning.keras')
-
-def predict_pokemon(image):
-    # Resize and preprocess the image
-    image = Image.fromarray((image * 255).astype(np.uint8))
-    image = image.resize((224, 224))
-    image_array = np.asarray(image)
-    image_array = image_array / 255.0
-
-
-    prediction = model.predict(np.expand_dims(image_array, axis=0))
-    predicted_class = np.argmax(prediction)
-
+labels = ['Haunter', 'Gengar', 'Ditto', 'Vulpix']
 
-    pokemon_names = ['Pikachu', 'Charmander', 'Bulbasaur', ...]
-    predicted_pokemon = pokemon_names[predicted_class]
+def predict_pokemon_type(uploaded_file):
+    if uploaded_file is None:
+        return "No file uploaded.", None, "No prediction"
 
-    return predicted_pokemon
+    model = tf.keras.models.load_model('pokemon-model_2_transferlearning.keras')
 
+    # Load the image from the file path
+    with Image.open(uploaded_file) as img:
+        img = img.resize((150, 150))
+        img_array = np.array(img)
 
-input_component = gr.inputs.Image(shape=(224, 224))
+        prediction = model.predict(np.expand_dims(img_array, axis=0))
+        confidences = {labels[i]: np.round(float(prediction[0][i]), 2) for i in range(len(labels))}
+        
+        # Identify the most confident prediction
+        confidences = {labels[i]: np.round(float(prediction[0][i]), 2) for i in range(len(labels))}
 
+        return img, confidences
 
-output_component = gr.outputs.Label(num_top_classes=1)
+# Define the Gradio interface
+iface = gr.Interface(
+    fn=predict_pokemon_type,  # Function to process the input
+    inputs=gr.File(label="Upload File"),  # File upload widget
+    outputs=["image", "text"],  # Output types for image and text
+    title="Pokemon Classifier",  # Title of the interface
+    description="Upload a picture of a Pokemon (preferably Cubone, Ditto, Psyduck, Snorlax, or Weedle) to see its type and confidence level. The trained model has an accuracy of 96%!"  # Description of the interface
+)
 
-gr.Interface(fn=predict_pokemon, inputs=input_component, outputs=output_component, title='Pokémon Classifier').launch()
+# Launch the interface
+iface.launch()