-
-
Notifications
You must be signed in to change notification settings - Fork 296
/
grad_cam.py
42 lines (35 loc) · 1.19 KB
/
grad_cam.py
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
"""
Simple implementation of Grad-CAM (https://arxiv.org/pdf/1610.02391.pdf)
"""
import numpy as np
from PIL import Image
import cv2
import dezero
import dezero.functions as F
from dezero.models import VGG16
url = 'https://github.com/oreilly-japan/deep-learning-from-scratch-3/raw/images/zebra.jpg'
img_path = dezero.utils.get_file(url)
img = Image.open(img_path)
img_size = img.size
model = VGG16(pretrained=True)
x = VGG16.preprocess(img)[np.newaxis] # preprocess for VGG
y = model(x)
last_conv_output = model.conv5_3.outputs[0]()
predict_id = np.argmax(y.data)
predict_output = y[0, predict_id]
predict_output.backward(retain_grad=True)
grads = last_conv_output.grad
pooled_grads = F.average(grads, axis=(0, 2, 3))
heatmap = last_conv_output.data[0]
for c in range(heatmap.shape[0]):
heatmap[c] *= pooled_grads[c].data
heatmap = np.mean(heatmap, axis=0)
heatmap = np.maximum(heatmap, 0)
heatmap /= np.max(heatmap)
# visualize the heatmap on image
img = cv2.imread(img_path)
heatmap = cv2.resize(heatmap, (img.shape[1], img.shape[0]))
heatmap = np.uint8(255 * heatmap)
heatmap = cv2.applyColorMap(heatmap, cv2.COLORMAP_JET)
heatmap_on_img = heatmap * 0.4 + img
cv2.imwrite('grad_cam.png', heatmap_on_img)