QA Official

relationship between ChannelPipeline and ChannelHandler in netty

https://qaofficial.com/post/2019/04/25/68717-relationship-between-channelpipeline-and-channelhandler-in-netty.html 2019-04-25
ChannelPipeline provides a container for the ChannelHandler chain and defines an API for propagating inbound and outbound event streams on the chain.When a Channel is created, it is automatically assigned to its own ChannelPipeline. ChannelHandler installation into ChannelPipeline: An implementation of ChannelInitializer is registered in ServerBootstrap (or Bootstrap for clients). when the ChannelInitializer.initChannel () method is called, ChannelInitializer will install a set of custom ChannelHandler in the ChannelPipeline; ChannelInitialier removes itself from ChannelPipeline.

sentiment analysis using Python's SnowNLP module

https://qaofficial.com/post/2019/04/25/68823-sentiment-analysis-using-pythonamp#x27s-snownlp-module.html 2019-04-25
SnowNLP is a python-written class library, which can conveniently process Chinese text content. It was written inspired by TextBlob. Since most natural language processing libraries are basically aimed at English, it has written a class library convenient to process Chinese. And unlike TextBlob, NLTK is not used here, all algorithms are implemented by itself, and some trained dictionaries are brought with it. Examples of calculating emotional values are as follows:

sqlpupper () function

https://qaofficial.com/post/2019/04/25/68647-sqlpupper-function.html 2019-04-25
sqlpupper () function instance code tutorial-returns the string str and changes all characters mapped according to the current character set to uppercase. example explanation: upper (field name): converts the contents of the field to uppercase for return note: the data type of the field must be a string type upper (string): Upper('aabbcc') returns' aabbcc' This method is seldom used in SQL Server, mostly in case-sensitive Oracle.

there are four labels, namely label1, label2, ..., label4, how to take the caption of the corresponding label according to an integer of 1-4

https://qaofficial.com/post/2019/04/25/68835-there-are-four-labels-namely-label1-label2-...-label4-how-to-take-the-caption-of-the-corresponding-label-according-to-an-integer-of-1-4.html 2019-04-25
there are four labels, namely label1, label2, ..., label4, how to take the caption traversal genus of the corresponding label according to an integer of 1-4  Method 1: Traverse all label controls on the owner. This traversal method stores the controls on a Tlist in the order in which they are placed on the owner. The controls are placed first and stored first procedure TForm1.Button1Click(Sender: TObject);var i: Integer;begin for I:=0 to Form1.

Ajax Post Request Solution to Chinese Scrambling Problem

https://qaofficial.com/post/2019/04/24/68557-ajax-post-request-solution-to-chinese-scrambling-problem.html 2019-04-24
This article introduces a solution to the problem of Chinese garbled code in ajax post. If you use utf-8 encoding in all your web applications, there will be no such problem.This is mainly to solve the problem of ajax post Chinese garbled code encountered by web applications that require page codes to be non-utf-8 codes such as gbk or gb2312 (sometimes due to problems left over from history). Generally, there are three solutions to this problem:

Ajax post request jump page

https://qaofficial.com/post/2019/04/24/68567-ajax-post-request-jump-page.html 2019-04-24
最近因为公司需求,需要ajax post请求并跳转界面,网上搜了一下资料,差不多都是用window.location.href来处理,但是在请

BN implementation

https://qaofficial.com/post/2019/04/24/68623-bn-implementation.html 2019-04-24
This should be the bn code closest to my understanding of the paper. Please correct me if you have any questions. def batch_norm(x, n_out,train, eps=1e-05, decay=0.99,affine=True, name=None): with tf.variable_scope(name, default_name='BatchNorm2d'): moving_mean = tf.get_variable('mean', [n_out], initializer=tf.zeros_initializer, trainable=False) moving_variance = tf.get_variable('variance', [n_out], initializer=tf.ones_initializer, trainable=False) train=tf.convert_to_tensor(train) def mean_var_with_update(): mean, variance = tf.nn.moments(x, [0,1,2], name='moments') # 计算train时的moving ave

MLC's Upper Page and Lower Page

https://qaofficial.com/post/2019/04/24/68627-mlcamp#x27s-upper-page-and-lower-page.html 2019-04-24
transferred from http://home.eeworld.com.cn/my/space-uid-346593-blogid-349102.html according to the different characteristics of Lower page and Upper page to solve different problems, it will bring excellent results in practical application, such as: improving performance, greatly improving life, solving the reliability and security risks of abnormal power failure, etc.Some special applications may bring other losses, such as: improving performance and life may require sacrificing capacity, etc., but these trade-offs are often more practical than common operations.

Netty Getting Started to Proficient Courses | Internet of Things Video Course

https://qaofficial.com/post/2019/04/24/68722-netty-getting-started-to-proficient-courses-internet-of-things-video-course.html 2019-04-24
《Netty Introduction to Core Technologies of Internet of Things to Proficient Courses》 本课程由www.51education.net独家提供 课程来源于: http://www.xuetuwuyou.com/course/14 课程内容介绍: 一、课程用

PCA Understanding Ideas

https://qaofficial.com/post/2019/04/24/68763-pca-understanding-ideas.html 2019-04-24
PCA has a lot of information on the Internet. People may read what Daniel wrote on the Internet about principles. I am only writing a basic understanding here.Mainly refer to the great god of information, https://zhuanlan.zhihu.com/p/21580949 PCA is mainly to find the eigenvector corresponding to the eigenvalue with the largest covariance matrix of the data set, thus finding the directions with the largest variance of the data, achieving the effect of dimension reduction for the data, reducing a vector with n dimensions to d dimensions, where d < n.