Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -3,10 +3,8 @@ import numpy as np
|
|
3 |
from keras.models import load_model
|
4 |
from PIL import Image
|
5 |
|
6 |
-
# Load your Keras model
|
7 |
model = load_model('pokemon-model_2_transferlearning.keras')
|
8 |
|
9 |
-
# Define function to preprocess and predict on images
|
10 |
def predict_pokemon(image):
|
11 |
# Resize and preprocess the image
|
12 |
image = Image.fromarray((image * 255).astype(np.uint8))
|
@@ -14,21 +12,20 @@ def predict_pokemon(image):
|
|
14 |
image_array = np.asarray(image)
|
15 |
image_array = image_array / 255.0
|
16 |
|
17 |
-
|
18 |
prediction = model.predict(np.expand_dims(image_array, axis=0))
|
19 |
predicted_class = np.argmax(prediction)
|
20 |
|
21 |
-
|
22 |
pokemon_names = ['Pikachu', 'Charmander', 'Bulbasaur', ...]
|
23 |
predicted_pokemon = pokemon_names[predicted_class]
|
24 |
|
25 |
return predicted_pokemon
|
26 |
|
27 |
-
|
28 |
input_component = gr.inputs.Image(shape=(224, 224))
|
29 |
|
30 |
-
|
31 |
output_component = gr.outputs.Label(num_top_classes=1)
|
32 |
|
33 |
-
# Create the Gradio interface
|
34 |
gr.Interface(fn=predict_pokemon, inputs=input_component, outputs=output_component, title='Pokémon Classifier').launch()
|
|
|
3 |
from keras.models import load_model
|
4 |
from PIL import Image
|
5 |
|
|
|
6 |
model = load_model('pokemon-model_2_transferlearning.keras')
|
7 |
|
|
|
8 |
def predict_pokemon(image):
|
9 |
# Resize and preprocess the image
|
10 |
image = Image.fromarray((image * 255).astype(np.uint8))
|
|
|
12 |
image_array = np.asarray(image)
|
13 |
image_array = image_array / 255.0
|
14 |
|
15 |
+
|
16 |
prediction = model.predict(np.expand_dims(image_array, axis=0))
|
17 |
predicted_class = np.argmax(prediction)
|
18 |
|
19 |
+
|
20 |
pokemon_names = ['Pikachu', 'Charmander', 'Bulbasaur', ...]
|
21 |
predicted_pokemon = pokemon_names[predicted_class]
|
22 |
|
23 |
return predicted_pokemon
|
24 |
|
25 |
+
|
26 |
input_component = gr.inputs.Image(shape=(224, 224))
|
27 |
|
28 |
+
|
29 |
output_component = gr.outputs.Label(num_top_classes=1)
|
30 |
|
|
|
31 |
gr.Interface(fn=predict_pokemon, inputs=input_component, outputs=output_component, title='Pokémon Classifier').launch()
|