标签:一个 mat training val site com image sts binary
原网址:https://data-flair.training/blogs/cnn-tensorflow-cifar-10/
by
2、卷积神经网络
在开始之前我们要充分理解CNN的原理。通常我们采用CIFAR 10数据集,这是一个物体识别的数据集,由60000张32*32的图片,包含10中类型的物体组成。
https://data-flair.training/blogs/wp-content/uploads/sites/2/2018/05/Typical_cnn.png
下载地址 https://www.cs.toronto.edu/~kriz/cifar.html.
3、CIFAR 10模型结构
该模型86%的正确率,需要在GPU上训练几小时。你需要下列文件:
cifar10_input.py Reads the native CIFAR-10 binary file format.
cifar10.py Builds the CIFAR-10 model.
cifar10_train.py Trains a CIFAR-10 model on a CPU or GPU.
cifar10_multi_gpu_train.py Trains a CIFAR-10 model on multiple GPUs.
cifar10_eval.py Evaluates the predictive performance of a CIFAR-10 model.
a.输入
下面还没翻译完 稍等
【翻译】TensorFlow卷积神经网络识别CIFAR 10Convolutional Neural Network (CNN)| CIFAR 10 TensorFlow
标签:一个 mat training val site com image sts binary
原文地址:https://www.cnblogs.com/zwczp/p/9945649.html