Create app.py

app.py

@@ -0,0 +1,40 @@
+# Downloading files from the demo repo
+import os
+os.mkdir('images') 
+!wget -q -O images/jokowi.jpeg https://cdn.setneg.go.id/_multimedia/photo/20220218/5008WhatsApp_Image_2022-02-18_at_1.36.50_PM.jpeg
+!wget -q -O images/megawati.jpeg https://gallery.poskota.co.id/storage/Foto/Foto_20220602_205953_hql.jpeg
+!wget -q -O images/cipung.jpg https://cdn.idntimes.com/content-images/community/2022/11/rayyanza-695b5fc766d9ed00ece029dcd8177b8e-4c74e93112d56ab97dac735945a7a619_600x400.jpg
+
+import gradio as gr
+from PIL import Image
+from transformers import pipeline
+
+# import the model
+pipe_age = pipeline("image-classification", model="nateraw/vit-age-classifier")
+pipe_emotion = pipeline("image-classification", model="ahyar002/emotion_classification")
+
+def age_prediction(image):
+    # convert to PIL image
+    pil_image = Image.fromarray(image)
+    # predict the image
+    predict_age = pipe_age(pil_image)
+    predict_emotion = pipe_emotion(pil_image)
+    # tranform the ouput into dictionary
+    transformed_dict_age = {item['label']: item['score'] for item in predict_age}
+    transformed_dict_emotion = {item['label']: item['score'] for item in predict_emotion}
+
+    return transformed_dict_age, transformed_dict_emotion
+
+
+demo = gr.Interface(age_prediction,
+    inputs = "image",
+    outputs= [gr.Label(num_top_classes=3), gr.Label(num_top_classes=3)],
+    examples=[
+        os.path.join(os.path.abspath(''), "images/jokowi.jpeg"),
+        os.path.join(os.path.abspath(''), "images/megawati.jpeg"),
+        os.path.join(os.path.abspath(''), "images/cipung.jpg"),
+    ],
+)
+
+if __name__ == "__main__":
+    demo.launch(debug=True)