Spaces:
Build error
Build error
taquynhnga
commited on
Commit
•
aff0cf3
1
Parent(s):
0343e34
Update pages/1_Maximally_activating_patches.py
Browse files
pages/1_Maximally_activating_patches.py
CHANGED
@@ -13,6 +13,15 @@ from backend.utils import load_dataset_dict
|
|
13 |
HIGHTLIGHT_COLOR = '#e7bcc5'
|
14 |
st.set_page_config(layout='wide')
|
15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
# -------------------------- LOAD DATASET ---------------------------------
|
17 |
dataset_dict = load_dataset_dict()
|
18 |
|
|
|
13 |
HIGHTLIGHT_COLOR = '#e7bcc5'
|
14 |
st.set_page_config(layout='wide')
|
15 |
|
16 |
+
|
17 |
+
st.title('Maximally activating image patches')
|
18 |
+
st.write('> What patterns maximally activate this channel in ConvNeXt model?')
|
19 |
+
st.write("""The maximally activating image patches method is a technique used in visualizing the interpretation of convolutional neural networks.
|
20 |
+
It works by identifying the regions of the input image that activate a particular neuron in the convolutional layer,
|
21 |
+
thus revealing the features that the neuron is detecting. To achieve this, the method generates image patches then feeds into the model while monitoring the neuron's activation.
|
22 |
+
The algorithm then selects the patch that produces the highest activation and overlays it on the original image to visualize the features that the neuron is responding to.
|
23 |
+
""")
|
24 |
+
|
25 |
# -------------------------- LOAD DATASET ---------------------------------
|
26 |
dataset_dict = load_dataset_dict()
|
27 |
|