Upload app.py

app.py

@@ -5,41 +5,42 @@ import gdown
 from PIL import Image
 
 
-
-input_shape = (32, 32, 3)
-resized_shape = (224, 224, 3)
-num_classes = 10
-labels = {
-    0: "plane",
-    1: "car",
-    2: "bird",
-    3: "cat",
-    4: "deer",
-    5: "dog",
-    6: "frog",
-    7: "horse",
-    8: "ship",
-    9: "truck",
-}
+labels = [
+    "plane",
+    "car",
+    "bird",
+    "cat",
+    "deer",
+    "dog",
+    "frog",
+    "horse",
+    "ship",
+    "truck",
+]
 
 # a file
 url = "https://drive.google.com/uc?id=12700bE-pomYKoVQ214VrpBoJ7akXcTpL"
 output = "modelV2Lmixed.keras"
 gdown.download(url, output, quiet=False)
 
-model = tf.keras.models.load_model("./modelV2Lmixed.keras")
+inception_net = tf.keras.models.load_model("./modelV2Lmixed.keras")
+
 
+def classify_image(inp):
+    inp = inp.reshape((-1, 224, 224, 3))
+    inp = tf.keras.applications.efficientnet.preprocess_input(inp)
+    prediction = inception_net.predict(inp).flatten()
+    confidences = {labels[i]: float(prediction[i]) for i in range(10)}
+    return confidences
 
-def classify_image(image, model):
-    image = tf.cast(image, tf.float32)
-    image = tf.image.resize(image, [32, 32])
-    image = np.expand_dims(image, axis=0)
-    prediction = model.predict(image)
-    #confidences = {labels[i]: float(prediction[i]) for i in range(10)}
-    return prediction
 
+import gradio as gr
 
-gr.Interface(fn=classify_image, 
-             inputs=gr.Image(shape=(32, 32)),
-             outputs=gr.Label(num_top_classes=3),
-             examples=["03_cat.jpg", "05_dog.jpg"]).launch()
+gr.Interface(
+    fn=classify_image,
+    inputs=gr.inputs.Image(shape=(32, 32)),
+    outputs=gr.outputs.Label(num_top_classes=3),
+    examples=["03_cat.jpg", "05_dog.jpg"],
+    theme="default",
+    css=".footer{display:none !important}",
+).launch()