机器学习实战源码

利用卷积神经网络进行特征提取，并将卷积特征可视化(Feature extraction using convolutional neural networks and visualization of convolution features)

bvmbvmnvnmm bmnmbbb

深度学习中，循环卷积网络常用于预测正弦函数的取值(Cyclic Convolution Networks for Predicting Sinusoidal Functions in Deep Learning)

说明：  python编写，使用sklearn机器学习库里的svm分类器，以MNIST为数据集，手动输入数字分类，MNIST.py为训练代码，训练10000次准确率为93.93%，手动3.py为手动输入并分类的代码。(Written in Python, using the svm classifier in the sklearn machine learning library, using MNIST as the data set, manually input the numeric classification, MNIST.py is the training code, the accuracy of training 10000 times is 93.93%, manual 3.py is manually entered and classified Code.)

时间序列python代码实现，运行调试时间序列python代码

说明：  针对序列信号分类识别Classification and recognition of sequence signals(Classification and recognition of sequence signals)

This example demonstrates how to read a raw file in Google-App-Engine. 1. Change name of application in app.yml to the name of you application 2. Upload the files 3. Open the main page

说明：  knn手写数字识别，tensorflow教程(Knn Handwritten Number Recognition)

说明：  数据挖掘作为一种现阶段很有用的方式，应用在各个领域。本文比较详细，可以借鉴看看。(Data mining, as a very useful way at this stage, is applied in various fields. We can learn from it.)

说明：  好学好用的机器学习经典算法实践供下载学习分享.(Good use of machine learning classic algorithm practice for download learning to share)