Update app.py

app.py

@@ -1,25 +1,21 @@
+from fastai.vision.all import *
 import gradio as gr
 
-# Assuming no shape needs to be specified directly in the constructor
-image = gr.Image()
-label = gr.Label()
-examples = [
-    '/kaggle/input/leukemia-cells/allbloodsmear.jpg', 
-    '/kaggle/input/cells-of-sickle/650x450-Sickle-Cell-Trait.jpg',
-    '/kaggle/input/cells-of-sickle/Sickle Cell Anemia smear 40x.jpg',
-    '/kaggle/input/healthy-cells/normalbloodsmear.jpg'
-]
+def get_label(x):
+    return 'dog' if 'dog' in x.name else 'cat'
+
+learn = load_learner('cat_or_dog.pkl')
+
+# Extract categories directly from the learner to ensure correct order
+categories = learn.dls.vocab
 
-intf = gr.Interface(
-    fn=classify_image, 
-    inputs=image, 
-    outputs=label, 
-    examples=examples, 
-    title='Blood Disease Identifier', 
-    description="Please upload your blood smear image that is greater than 200x magnification to diagnose the patient with the following blood diseases: Sickle Cell Disease, Leukemia, or Thalassemia. <br><br>Please note that the results are not 100% accurate and should only be used as a first means of detection. Please contact a medical professional for more information and possible future action. HematoTech are not liable for any misdiagnosis that may occur.", 
-    allow_flagging=False, 
-    article='Sickle cell disease (SCD) is a group of inherited red blood cell disorders. In SCD, the red blood cells become hard and sticky and look like a C-shaped farm tool called a “sickle.” The disease can be managed under proper medical supervision.'
-)
+def classify_img(img):
+    pred, idx, probs = learn.predict(img)
+    return {categories[i]: float(probs[i]) for i in range(len(categories))}
+
+image = gr.Image()
+labels = gr.Label()
+examples = ['dog.jpg', 'cat.jpeg']
 
-# Launch the interface
-intf.launch(inline=False)
+intf = gr.Interface(fn=classify_img, inputs=image, outputs=labels, examples=examples)
+intf.launch(inline=False, share=True)