人脸识别python代码

基于模拟退火算法对BP神经网络进行的改进，有效提高了精度和迭代速度。(The improvement of BP neural network based on simulated annealing algorithm effectively improves the accuracy and speed of iteration.)

对0~9的数字进行识别训练。利用tensorflow构建卷积神经网络的框架。(Identify training for 0~9 numbers)

深度学习算法 主要是GDAL处理栅格数据转换为矢量数据 利用深度学习的算法实现(In-depth learning algorithm is mainly realized by GDAL in converting raster data into vector data and using in-depth learning algorithm.)

给定一个训练数据集，对新的输入实例，在训练数据集中找到与该实例最邻近的 k （k <= 20）个实例，这 k 个实例的多数属于某个类， 就把该输入实例分为这个类。(Given a training data set, the nearest K (k <= 20) instances to the new input instance are found in the training data set. Most of the K instances belong to a class. The input instance is categorized into this class.)

说明：  lstm时间序列的python语言，可以参考(Python language of LSTM time series)

说明：  从入门到实践，比较好的适宜入门的Python类书籍(From entry to practice)

Python开发计算器-多功能版，精简版1.0可以进入我的主页查找

LOF（局部异常因子）是用于识别基于密度的局部异常值的算法(It uses local outlier mining method to count the Local Outlier Factor(LOF) of the outlier candidated object。)

machine learning 方面有关于Classification fundamentals的资料，语言是python(Classification fundamentals aspects of machine learning have information about language is python)

说明：  利用GPflow进行高斯过程回归，kernel有OU过程核等(Gauss Process Regression Using GPflow)