caffe python接口：可视化每层图像特征

原文链接

import numpy as py
import matplotlib.pyplot as plt
import sys,os
import caffe

#设置当前目录
caffe_root="/home/XXX/caffe/"   #caffe所在路径
os.chdir(caffe_root)
sys.path.insert(0,caffe_root+'python')

#显示图形大小为10*10，图形的插值以最近为原则，图像的颜色是灰色
plt.rcParams['figure.figsize'] = (10, 10)
plt.rcParams['image.interpolation'] = 'nearest'
plt.rcParams['image.cmap'] = 'gray'


#利用提前训练好的模型，设置测试网络
caffe.set_mode_gpu()
net = caffe.Net(caffe_root + 'examples/cifar10/cifar10_quick.prototxt',
                caffe_root + 'examples/cifar10/cifar10_quick_iter_4000.caffemodel',
                caffe.TEST)


#数据层的的shape
print net.blobs['data'].data.shape  #（1，3，32，32）

#加载测试图片，并显示
im = caffe.io.load_image('examples/images/32.jpg')
print im.shape  #图片shape(32,32,3)
plt.imshow(im)
plt.axis('off')


#　编写一个函数，将二进制的均值转换为python的均值
def convert_mean(binMean,npyMean):
    blob = caffe.proto.caffe_pb2.BlobProto()
    bin_mean = open(binMean, 'rb' ).read()
    blob.ParseFromString(bin_mean)
    arr = np.array( caffe.io.blobproto_to_array(blob) )
    npy_mean = arr[0]
    np.save(npyMean, npy_mean )
binMean=caffe_root+'examples/cifar10/mean.binaryproto'
npyMean=caffe_root+'examples/cifar10/mean.npy'
convert_mean(binMean,npyMean)


#将图片载入blob中,并减去均值
transformer = caffe.io.Transformer({'data': net.blobs['data'].data.shape})
transformer.set_transpose('data', (2,0,1))
transformer.set_mean('data', np.load(npyMean).mean(1).mean(1)) # 减去均值
transformer.set_raw_scale('data', 255)  
transformer.set_channel_swap('data', (2,1,0))
net.blobs['data'].data[...] = transformer.preprocess('data',im)
inputData=net.blobs['data'].data

#显示减去均值前后的数据
plt.figure()
plt.subplot(1,2,1),plt.title("origin")
plt.imshow(im)
plt.axis('off')
plt.subplot(1,2,2),plt.title("subtract mean")
plt.imshow(transformer.deprocess('data', inputData[0]))
plt.axis('off')


运行测试模型，并显示各层数据信息
net.forward()
print [(k, v.data.shape) for k, v in net.blobs.items()]
#输出如下：
[('data', (1, 3, 32, 32)),
 ('conv1', (1, 32, 32, 32)),
 ('pool1', (1, 32, 16, 16)),
 ('conv2', (1, 32, 16, 16)),
 ('pool2', (1, 32, 8, 8)),
 ('conv3', (1, 64, 8, 8)),
 ('pool3', (1, 64, 4, 4)),
 ('ip1', (1, 64)),
 ('ip2', (1, 10)),
 ('prob', (1, 10))]

#显示各层的参数信息v[0]权重,v[1] bias
print [(k, v[0].data.shape) for k, v in net.params.items()]
#输出如下：
[('conv1', (32, 3, 5, 5)),
 ('conv2', (32, 32, 5, 5)),
 ('conv3', (64, 32, 5, 5)),
 ('ip1', (64, 1024)),
 ('ip2', (10, 64))]


#编写一个函数，用于显示各层数据
def show_data(data, padsize=1, padval=0):
    data -= data.min()
    data /= data.max()

    # 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, padsize), (0, padsize)) + ((0, 0),) * (data.ndim - 3)
    data = np.pad(data, padding, mode='constant', constant_values=(padval, padval))

    # 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.figure()
    plt.imshow(data,cmap='gray')
    plt.axis('off')
plt.rcParams['figure.figsize'] = (8, 8)
plt.rcParams['image.interpolation'] = 'nearest'
plt.rcParams['image.cmap'] = 'gray'


#显示第一个卷积层的输出数据和权值（filter）
show_data(net.blobs['conv1'].data[0])
print net.blobs['conv1'].data.shape
show_data(net.params['conv1'][0].data.reshape(32*3,5,5))
print net.params['conv1'][0].data.shape

#fc层输出的直方图分布
feat=net.blobs['fc6'].data[0]
plt.subplot(2,1,1)
plt.plot(feat.flat)
plt.plot(2,1,2)
_=plt.hist(feat.flat[]feat.flat>0],bins=100)
![这里写图片描述](https://img-blog.csdn.net/20171120215925685?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvcXFfMzQ2Mzc0MDg=/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)

# 最后一层输入属于某个类的概率
feat = net.blobs['prob'].data[0]
print feat
plt.plot(feat.flat)  #绘制概率类别图

#输出如下：
[  5.21440245e-03   1.58397834e-05   3.71246301e-02   2.28459597e-01
   1.08315737e-03   7.17785358e-01   1.91939052e-03   7.67927198e-03
   6.13298907e-04   1.05107691e-04]