File size: 3,502 Bytes
839b7e0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
# Importing some modules
import gradio as gr
from transformers import pipeline
import torch.cuda as cuda

# Loading in the model
MODEL = pipeline('image-classification', model='nateraw/vit-age-classifier', device=0 if cuda.is_available() else -1)

# Main function to classify image
def classify_image(image, top_k):
    # Getting the classification result
    classification_result = MODEL(image)

    # Reformating the classification result into a dictionary
    classification_result = {result['label']: result['score'] for result in classification_result[:int(top_k)]}

    # Add some text comment to it lol
    comment = text_comment(list(classification_result.keys())[0])

    # Returning the classification result
    return classification_result, comment

# Snarky comment based on age
def text_comment(pred_class):
    match pred_class:
        case "3-9":
            return "Lost your way to the playground?"
        case "10-19":
            return "But Mom, I'm not a kid anymore!"
        case "20-29":
            return "You're in your prime!"
        case "30-39":
            return "Oof, watch out for those wrinkles!"
        case "40-49":
            return "You're still young at heart!"
        case "50-59":
            return "Retirement is just around the corner!"
        case "60-69":
            return "You're a senior citizen now!"
        case "more than 70":
            return "Hey Siri, play 'My Way' by Frank Sinatra"
        

if __name__ == "__main__":
    # Creating the Gradio interface
    with gr.Blocks() as demo:
        gr.Markdown("""
                    # I will guess your age based on your picture!
                    ---
                    Totally not creepy, I promise :)
                    <br>Made by [Dennis Jonathan](dennisjooo.github.io). A project for REA Mastering AI course.   
                    Age guessing model from [nateraw/vit-age-classifier](https://huggingface.co/nateraw/vit-age-classifier)
                    """)
        
        with gr.Row(equal_height=True):
            with gr.Column():
                # Creating the input block
                image = gr.Image(label="Upload a picture of yourself", type="pil", scale=2)

                # Creating the example block
                gr.Examples(examples=[
                    "./images/andrew.jpg",
                    "./images/feifei.jpg",
                    "./images/geoff.jpg",
                    "./images/ilya.jpg",
                    "./images/karpathy.jpg",
                    "./images/lex.jpg"
                    ], inputs=[image], label="Or choose an example")
        

            with gr.Column():
                # Getting the top k hyperparameter
                top_k = gr.Number(label="How many guesses do I get?", value=1)

                # Creating the output block
                label = gr.Label(label="Hey it's me, your age!")
                comment = gr.Textbox(label="Based on your age, I think you are...", 
                                    placeholder="I'm still learning, so I might be wrong!")

        with gr.Row():
            # Submit button
            btn = gr.Button("Beep boop, guess my age!")
            btn.click(classify_image, inputs=[image, top_k], outputs=[label, comment])

            # Clear button
            clear = gr.Button("Poof begone!")
            clear.click(lambda: [None, None, None], inputs=[], outputs=[image, label, comment])

    # Launching the interface
    demo.launch(share=False, debug=True)