caffe画 feature map

落日映苍穹つ 2022-06-03 01:12 102阅读 0赞

import numpy as np  
import matplotlib.pyplot as plt  
import sys  
import os  
import caffe  
plt.rcParams\['figure.figsize'\] = (10, 10) \# large images  
plt.rcParams\['image.interpolation'\] = 'nearest' \# don't interpolate: show square pixels  
plt.rcParams\['image.cmap'\] = 'gray' \# use grayscale output rather than a (potentially misleading) color heatmap  
  
  
caffe\_root = 'D:/\*/caffe-master/' \# this file should be run from \{caffe\_root\}/examples (otherwise change this line)  
model\_root = 'D:/RFCN/'  
  
  
sys.path.insert(0, caffe\_root + 'python')  
  
  
caffe.set\_mode\_cpu()  
  
  
model\_def = model.prototxt'  
model\_weights = model\_root + 'resnet50\_iter\_110000.caffemodel'  
  
  
net = caffe.Net(model\_def,model\_weights,caffe.TEST)  
transformer = caffe.io.Transformer(\{'data': net.blobs\['data'\].data.shape\})  
transformer.set\_transpose('data', (2,0,1))  
transformer.set\_raw\_scale('data', 255)  
transformer.set\_channel\_swap('data', (2,1,0)) \# swap channels from RGB to BGR  
print net.blobs\['data'\].data.shape  
image = caffe.io.load\_image('D:/000001.png')  
transformed\_image = transformer.preprocess('data', image)  
plt.imshow(image)  
plt.show()  
  
  
net.blobs\['data'\].data\[...\] = transformed\_image  
output = net.forward()  
  
  
def vis\_square(data):  
"""Take an array of shape (n, height, width) or (n, height, width, 3)  
and visualize each (height, width) thing in a grid of size approx. sqrt(n) by sqrt(n)"""  
  
  
\# normalize data for display  
data = (data - data.min()) / (data.max() - data.min())  
  
  
\# force the number of filters to be square  
n = int(np.ceil(np.sqrt(data.shape\[0\])))  
padding = (((0, n \*\* 2 - data.shape\[0\]),  
(0, 1), (0, 1)) \# add some space between filters  
\+ ((0, 0),) \* (data.ndim - 3)) \# don't pad the last dimension (if there is one)  
data = np.pad(data, padding, mode='constant', constant\_values=1) \# pad with ones (white)  
  
  
\# tile the filters into an image  
data = data.reshape((n, n) + data.shape\[1:\]).transpose((0, 2, 1, 3) + tuple(range(4, data.ndim + 1)))  
data = data.reshape((n \* data.shape\[1\], n \* data.shape\[3\]) + data.shape\[4:\])  
  
  
plt.imshow(data)  
plt.axis('off')  
plt.show()  
  
  
filters = net.blobs\['ipmdata'\].data  
print(filters.shape)  
vis\_square(filters.transpose(0, 2, 3, 1))