Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,4 +1,3 @@
|
|
1 |
-
import cv2
|
2 |
import gradio as gr
|
3 |
import numpy as np
|
4 |
import tensorflow as tf
|
@@ -14,13 +13,9 @@ _RESOLUTION = 224
|
|
14 |
def get_model() -> tf.keras.Model:
|
15 |
"""Initiates a tf.keras.Model from HF Hub."""
|
16 |
inputs = tf.keras.Input((_RESOLUTION, _RESOLUTION, 3))
|
17 |
-
hub_module = from_pretrained_keras(
|
18 |
-
"probing-vits/cait_xxs24_224_classification"
|
19 |
-
)
|
20 |
|
21 |
-
logits, sa_atn_score_dict, ca_atn_score_dict = hub_module(
|
22 |
-
inputs, training=False
|
23 |
-
)
|
24 |
|
25 |
return tf.keras.Model(
|
26 |
inputs, [logits, sa_atn_score_dict, ca_atn_score_dict]
|
@@ -38,17 +33,15 @@ def show_plot(image):
|
|
38 |
_, _, ca_atn_score_dict = _MODEL.predict(preprocessed_image)
|
39 |
|
40 |
# Compute the saliency map and superimpose.
|
41 |
-
|
42 |
preprocessed_image, ca_atn_score_dict, block_key="ca_ffn_block_0_att"
|
43 |
)
|
44 |
-
|
45 |
-
|
46 |
-
)
|
47 |
-
|
|
|
48 |
|
49 |
-
saliency_map = heatmap + original_image
|
50 |
-
saliency_map = np.clip(saliency_map, 0.0, 255.0).astype(np.uint8)
|
51 |
-
return Image.fromarray(saliency_map)
|
52 |
|
53 |
|
54 |
title = "Generate Class Saliency Plots"
|
@@ -57,10 +50,10 @@ article = "Class saliency maps as investigated in [Going deeper with Image Trans
|
|
57 |
iface = gr.Interface(
|
58 |
show_plot,
|
59 |
inputs=gr.inputs.Image(type="pil", label="Input Image"),
|
60 |
-
outputs="
|
61 |
title=title,
|
62 |
article=article,
|
63 |
allow_flagging="never",
|
64 |
-
examples=[["./
|
65 |
)
|
66 |
-
iface.launch()
|
|
|
|
|
1 |
import gradio as gr
|
2 |
import numpy as np
|
3 |
import tensorflow as tf
|
|
|
13 |
def get_model() -> tf.keras.Model:
|
14 |
"""Initiates a tf.keras.Model from HF Hub."""
|
15 |
inputs = tf.keras.Input((_RESOLUTION, _RESOLUTION, 3))
|
16 |
+
hub_module = from_pretrained_keras("probing-vits/cait_xxs24_224_classification")
|
|
|
|
|
17 |
|
18 |
+
logits, sa_atn_score_dict, ca_atn_score_dict = hub_module(inputs, training=False)
|
|
|
|
|
19 |
|
20 |
return tf.keras.Model(
|
21 |
inputs, [logits, sa_atn_score_dict, ca_atn_score_dict]
|
|
|
33 |
_, _, ca_atn_score_dict = _MODEL.predict(preprocessed_image)
|
34 |
|
35 |
# Compute the saliency map and superimpose.
|
36 |
+
saliency_attention = utils.get_cls_attention_map(
|
37 |
preprocessed_image, ca_atn_score_dict, block_key="ca_ffn_block_0_att"
|
38 |
)
|
39 |
+
fig = plt.figure()
|
40 |
+
plt.imshow(original_image.astype("int32"))
|
41 |
+
plt.imshow(saliency_attention.squeeze(), cmap="cividis", alpha=0.9)
|
42 |
+
plt.axis("off")
|
43 |
+
return fig
|
44 |
|
|
|
|
|
|
|
45 |
|
46 |
|
47 |
title = "Generate Class Saliency Plots"
|
|
|
50 |
iface = gr.Interface(
|
51 |
show_plot,
|
52 |
inputs=gr.inputs.Image(type="pil", label="Input Image"),
|
53 |
+
outputs=gr.outputs.Plot(type="auto"),
|
54 |
title=title,
|
55 |
article=article,
|
56 |
allow_flagging="never",
|
57 |
+
examples=[["./butterfly_cropped.png"]],
|
58 |
)
|
59 |
+
iface.launch(debug=True)
|