Update app.py

app.py

@@ -1,43 +1,26 @@
+
+__all__ = ['is_katy', 'learn', 'classify_image', 'categories', 'image', 'label', 'examples', 'intf']
+
+# Cell
+from fastai.vision.all import *
 import gradio as gr
-import pickle
-import numpy as np
-from PIL import Image
-import torch
-from torchvision import transforms
-
-# Load the pre-trained model
-with open('model.pkl', 'rb') as f:
-    model = pickle.load(f)
-
-# Define the image transformations
-transform = transforms.Compose([
-    transforms.Resize((224, 224)),
-    transforms.ToTensor(),
-    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
-])
-
-# Define the prediction function
-def predict(image):
-    # Apply the transformations to the input image
-    img = transform(image).unsqueeze(0)
-    model.eval()
-    with torch.no_grad():
-        output = model(img)
-        prediction = torch.argmax(output, dim=1).item()
-        
-    # Define class names
-    class_names = ["Not Katy Perry", "Katy Perry"]
-    
-    return class_names[prediction]
-
-# Set up the Gradio interface
-iface = gr.Interface(
-    fn=predict,
-    inputs=gr.inputs.Image(type="pil"),
-    outputs="text",
-    title="Katy Perry Recognition",
-    description="Upload an image to find out if it's Katy Perry or not.",
-)
-
-if __name__ == "__main__":
-    iface.launch()
+
+def is_katy(x): return x[0].isupper()
+
+# Cell
+learn = load_learner('model.pkl')
+
+# Cell
+categories = ('Katy', 'Zooey')
+
+def classify_image(img):
+    pred,idx,probs = learn.predict(img)
+    return dict(zip(categories, map(float,probs)))
+
+# Cell
+image = gr.inputs.Image(shape=(192, 192))
+label = gr.outputs.Label()
+examples = ['katy.jpg', 'katy2.jpg','zooey1.jpg','zooey2.jpg','zooey3.jpg','dunno.jpg']
+
+intf = gr.Interface(fn=classify_image, inputs=image, outputs=label, examples=examples)
+intf.launch(inline=False)