TensorFlow创建简单的图片分类系统--机器学习

TensorFlow

参考链接 http://www.wolfib.com/Image-Recognition-Intro-Part-1/

环境要求

linux amd64（必须是64位处理器）
python 2.7
TensorFlow

环境安装

···
//安装Python-pip 和python-dev
sudo apt-get install python-pip python-dev

//安装TensorFlow模块
sudo pip install --upgrade https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.8.0-cp27-none-linux_x86_64.whl

···

资源

https://www.cs.toronto.edu/~kriz/cifar.html

拷贝到项目根目录下

源码

https://github.com/wolfib/image-classification-CIFAR10-tf

运行结果

lc@ubuntu:~/Desktop/py/learn$ python softmax.py
Step     0: training accuracy 0.08
Step   100: training accuracy 0.26
Step   200: training accuracy 0.29
Step   300: training accuracy 0.2
Step   400: training accuracy 0.35
Step   500: training accuracy 0.34
Step   600: training accuracy 0.37
Step   700: training accuracy 0.27
Step   800: training accuracy 0.28
Step   900: training accuracy 0.38
Test accuracy 0.2179
Total time: 26.82s