Update app.py

app.py

@@ -1,3 +1,55 @@
+# import gradio as gr
+# import tensorflow as tf
+# import numpy as np
+# import os
+
+# # Configuration
+# HEIGHT, WIDTH = 224, 224
+# NUM_CLASSES = 6
+# LABELS = [ "McDonalds","Burger King","Subway", "Starbucks", "KFC","Other"]
+# from tensorflow_addons.metrics import F1Score
+# from keras.utils import custom_object_scope
+
+# with custom_object_scope({'Addons>F1Score': F1Score}):
+#     model = tf.keras.models.load_model('best_model.h5')
+
+
+# def classify_image(inp):
+#     np.random.seed(143)
+
+#     # Ensure input is resized to expected shape
+#     inp = tf.image.resize(inp, [HEIGHT, WIDTH])
+#     inp = tf.cast(inp, tf.float32)  # ensure correct dtype for preprocessing
+#     inp = tf.keras.applications.nasnet.preprocess_input(inp)
+#     inp = tf.expand_dims(inp, axis=0)  # make batch dimension
+
+#     # Prediction
+#     prediction = model.predict(inp)
+
+#     return {LABELS[i]: float(f"{prediction[0][i]:.6f}") for i in range(NUM_CLASSES)}
+
+
+
+
+# iface = gr.Interface(
+#     fn=classify_image,
+#     inputs=gr.Image(
+#         label="Input Image",
+#         source="upload",
+#         type="numpy",
+#         height=HEIGHT,
+#         width=WIDTH
+#     ),
+#     outputs=gr.Label(num_top_classes=4),
+#     title="Brand Logo Detection",
+#     examples=example_list
+# )
+
+# if __name__ == "__main__":
+#     iface.launch(debug=False,share=True)
+
+
+
 import gradio as gr
 import tensorflow as tf
 import numpy as np
@@ -6,7 +58,8 @@ import os
 # Configuration
 HEIGHT, WIDTH = 224, 224
 NUM_CLASSES = 6
-LABELS = [ "McDonalds","Burger King","Subway", "Starbucks", "KFC","Other"]
+LABELS = ["McDonalds", "Burger King", "Subway", "Starbucks", "KFC", "Other"]
+
 from tensorflow_addons.metrics import F1Score
 from keras.utils import custom_object_scope
 
@@ -15,19 +68,15 @@ with custom_object_scope({'Addons>F1Score': F1Score}):
 
 
 def classify_image(inp):
-    np.random.seed(143)
-
-    # Ensure input is resized to expected shape
+    # Resize & preprocess
     inp = tf.image.resize(inp, [HEIGHT, WIDTH])
-    inp = tf.cast(inp, tf.float32)  # ensure correct dtype for preprocessing
+    inp = tf.cast(inp, tf.float32)
     inp = tf.keras.applications.nasnet.preprocess_input(inp)
-    inp = tf.expand_dims(inp, axis=0)  # make batch dimension
-
-    # Prediction
-    prediction = model.predict(inp)
-
-    return {LABELS[i]: float(f"{prediction[0][i]:.6f}") for i in range(NUM_CLASSES)}
+    inp = tf.expand_dims(inp, axis=0)
 
+    # Predict
+    prediction = model.predict(inp)[0]
+    return {LABELS[i]: float(f"{prediction[i]:.6f}") for i in range(NUM_CLASSES)}
 
 example_list = [
     ["Examples/Untitled.png"],
@@ -36,6 +85,7 @@ example_list = [
     ["Examples/Untitled5.png"]
 ]
 
+
 iface = gr.Interface(
     fn=classify_image,
     inputs=gr.Image(
@@ -51,4 +101,5 @@ iface = gr.Interface(
 )
 
 if __name__ == "__main__":
-    iface.launch(debug=False,share=True)
+    iface.launch(debug=False, share=True)
+