这个程序是去南京进行机器人跳舞比赛用的程序

说明：  python 实现的bp神经网络预测，数据已经有，可以直接替换，叠加次数还有步长可以自己设置。(Python implementation of BP neural network prediction, the data already has, can be directly replaced, the number of overlap, there are steps you can set yourself.)

图像切割源码，经典必备，图像识别参考

说明：  LSTM模型进行股票预测，时间序列预测，LSTM算法(300ETF stock predict by LSTM model)

Comport adjustment for winfof

说明：  训练一个卷积神经网络，用fastai库（建在PyTorch上）将图像分类为纸板，玻璃，金属，纸张，塑料或垃圾。数据集下载地址如下。(Train a convolutional neural network to classify images into cardboard, glass, metal, paper, plastic or garbage using the fastai Library (built on pytorch). The data set download address is as follows.)

利用VORONOI算法在PYTHON中的多晶材料建模(Modeling of Polycrystalline Materials in PYTHON Using VORONOI Algorithms)

利用LSTM 生成时间曲线， 然后利用实际值与预测值进行比较，查看结果好坏(use LSTM generate new time series, and compare the prediction with target to see if the result is good or not.)

微信跳一跳自动化代码，使用python语言编写，需要手机进入开发者模式

说明：  基于dlib库的人脸68个特征点训练模型的扩展模型，可识别出人脸81个特征点，包括额头部分。(Based on the extended model of the face training model of 68 feature points based on Dlib database, 81 feature points can be recognized, including the forehead part.)

Python脚本。模拟登录知乎， 爬虫，操作excel，微信公众号，远程开机(Python script. Simulated login, crawler, excel, Wechat public number, remote boot)