caffe前向传播

马列之

于 2019-06-17 10:22:56 发布

阅读量427

点赞数

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/aboy2018/article/details/92575332

博客围绕Caffe前向传播展开，虽未提供具体内容，但可知聚焦于Caffe框架的前向传播这一关键技术环节，在深度学习领域有重要意义。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

import os
import sys
import numpy as np


sys.path.insert(0,"/home/jiashuaihe/Downloads/caffe/python")

deploy = "deploy.prototxt"
caffe_model = "./model/pred_iter_3260.caffemodel"
img1 = "./test_resize/259665.png"
labels_filename = 'train_test.txt'
os.environ["GLOG_alsologtostderr"] = "1"
import caffe
#import google.protobuf.text_format 
#caffe.caffe_init_glog()
import google.protobuf as pb2
# the basic gpu config
caffe.set_device(1)
caffe.set_mode_gpu()


net = caffe.Net(deploy, caffe_model, 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)

net.blobs['data'].reshape(50,3,224,224)
im2 = caffe.io.load_image(img1,False)
print(im2.shape)
net.blobs['data'].data[...] = transformer.preprocess('data',im2)
print(net.blobs['data'].shape)
net.forward()

labels = np.loadtxt(labels_filename, str, delimiter='\t')
prob= net.blobs['prob'].data[0].flatten()
order=prob.argsort()[-1]

print('the class is:',labels[order])