File size: 1,730 Bytes
ed1b9ba
 
 
3281fd3
b0daa79
ed1b9ba
 
 
 
 
 
 
 
 
 
 
cdee39b
ed1b9ba
cdee39b
ed1b9ba
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cdee39b
1d20657
ed1b9ba
cdee39b
 
 
 
 
ed1b9ba
 
 
 
 
 
 
 
 
cdee39b
ed1b9ba
 
 
cdee39b
ed1b9ba
cdee39b
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
import gradio as gr
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt

from huggingface_hub.keras_mixin import from_pretrained_keras
from PIL import Image

import utils

_RESOLUTION = 224


def get_model() -> tf.keras.Model:
    """Initiates a tf.keras.Model from HF Hub."""
    inputs = tf.keras.Input((_RESOLUTION, _RESOLUTION, 3))
    hub_module = from_pretrained_keras("probing-vits/cait_xxs24_224_classification")

    logits, sa_atn_score_dict, ca_atn_score_dict = hub_module(inputs, training=False)

    return tf.keras.Model(
        inputs, [logits, sa_atn_score_dict, ca_atn_score_dict]
    )


_MODEL = get_model()


def show_plot(image):
    """Function to be called when user hits submit on the UI."""
    original_image, preprocessed_image = utils.preprocess_image(
        image, _RESOLUTION
    )
    _, _, ca_atn_score_dict = _MODEL.predict(preprocessed_image)

    # Compute the saliency map and superimpose.
    saliency_attention = utils.get_cls_attention_map(
        preprocessed_image, ca_atn_score_dict, block_key="ca_ffn_block_0_att"
    )
    fig = plt.figure()
    plt.imshow(original_image.astype("int32"))
    plt.imshow(saliency_attention.squeeze(), cmap="cividis", alpha=0.9)
    plt.axis("off")
    return fig



title = "Generate Class Saliency Plots"
article = "Class saliency maps as investigated in [Going deeper with Image Transformers](https://arxiv.org/abs/2103.17239) (Touvron et al.)."

iface = gr.Interface(
    show_plot,
    inputs=gr.inputs.Image(type="pil", label="Input Image"),
    outputs=gr.outputs.Plot(type="auto"),
    title=title,
    article=article,
    allow_flagging="never",
    examples=[["./butterfly_cropped.png"]],
)
iface.launch(debug=True)