Caffe各层可视化, python实现

最新推荐文章于 2020-01-12 22:16:56 发布

吃我老猪一钉耙

最新推荐文章于 2020-01-12 22:16:56 发布

阅读量312

点赞数

分类专栏：深度学习caffe-pycaffe 文章标签： Caffe各层可视化 python

本文链接：https://blog.csdn.net/zlf19910726/article/details/78699756

版权

深度学习caffe-pycaffe 专栏收录该内容

9 篇文章 0 订阅

订阅专栏

#!/usr/bin/env python

import numpy as np
import matplotlib.pyplot as plt
#%matplotlib inline
import sys,os,caffe

################################################################################

caffe_root = '/home/zlf/my_software/my_caffe_configure/caffe-master/'

sys.path.insert(0, caffe_root + 'python')
os.chdir(caffe_root)
if not os.path.isfile(caffe_root + 'examples/ScatterParamsOld/cifar10_quick_iter_10000.caffemodel'):
    print "caffemodel is not exist..."
else:
    print "Ready to Go ..."
caffe.set_mode_gpu()

net = caffe.Net(caffe_root + 'examples/ScatterParamsOld/deploy.protxt',
                caffe_root + 'examples/ScatterParamsOld/cifar10_quick_iter_10000.caffemodel',
                caffe.TEST)

print(net.blobs['data'].data.shape)

im = caffe.io.load_image('/home/zlf/Documents/LC20816B_A11k1_1k2_1p1_1p2_1.tif')





















print(im.shape)
plt.imshow(im)
plt.axis('off')

################################################################################

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)
#print(np.load(npyMean))
print("np.load(npyMean).shape")
print(np.load(npyMean).shape)

################################################################################

#import pdb
#pdb.set_trace()

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)) 
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

################################################################################

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')

################################################################################

net.forward()
for layer_name, blob in net.blobs.iteritems():
    print layer_name + '\t' + str(blob.data.shape)


print("\n\n")
for layer_name, param in net.params.iteritems():
    print layer_name + '\t' + str(param[0].data.shape), str(param[1].data.shape)


################################################################################

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'

################################################################################

show_data(net.blobs['conv1'].data[0])
print ("net.blobs['conv1'].data.shape:" +str(net.blobs['conv1'].data.shape) )


#show_data(net.params['conv1'][0].data.reshape(32*3,5,5))
#print ("net.params['conv1'][0].data.shape:" +str(net.params['conv1'][0].data.shape) )


show_data(net.blobs['pool1'].data[0])
print ("net.blobs['pool1'].data.shape:" +str(net.blobs['pool1'].data.shape) )


show_data(net.blobs['conv2'].data[0],padval=0.5)
print ("net.blobs['conv2'].data.shape:" +str(net.blobs['conv2'].data.shape) )


#show_data(net.params['conv2'][0].data.reshape(32*2,5,5))
#print ("net.params['conv2'][0].data.shape" +str(net.params['conv2'][0].data.shape) )


show_data(net.blobs['conv3'].data[0],padval=0.5)
print ("net.blobs['conv3'].data.shape:" +str(net.blobs['conv3'].data.shape) )


#show_data(net.params['conv3'][0].data.reshape(64*32,5,5)[:1024])
#print ("net.params['conv3'][0].data.shape:" +str(net.params['conv3'][0].data.shape) )

#show_data(net.blobs['pool3'].data[0],padval=0.2)
#print ("net.blobs['pool3'].data.shape:" +str(net.blobs['pool3'].data.shape) )


feat = net.blobs['prob'].data[0]
print feat
plt.figure()
plt.plot(feat.flat)