QA Official

single classification

https://qaofficial.com/post/2019/04/20/68377-single-classification.html 2019-04-20
Single classification algorithms used in engineering practice include one-class svm and svdd, both of which have good effects, but one-class svm has slightly better effects than svdd, but svdd has fewer support vectors, which can achieve better real-time performance in practice. iforest is also another single classification algorithm, but it has not been tried yet. I don't know how the actual effect is, but it needs to be tried.

spring Startup Process-Principles of Spring and springMVC Parent-Child Containers

https://qaofficial.com/post/2019/04/20/68322-spring-startup-process-principles-of-spring-and-springmvc-parent-child-containers.html 2019-04-20
To understand the relationship between these three contexts, you need to be familiar with how spring is started in a web container.Spring's startup process is actually the startup process of its IoC container. For web programs, the startup process of IoC container is the process of establishing context. spring startup process: First of all, for a web application, it is deployed in a web container, which provides a global context environment.

sys.argv usage

https://qaofficial.com/post/2019/04/20/68422-sys.argv-usage.html 2019-04-20
The sys.argv variable is a list of strings.Contains a list of command line arguments, that is, arguments passed to your program using the command line.When using Python to perform command line operations, the contents behind the script are passed to the program as parameters.Python stores parameters in sys.argv variables for us.Remember: the name of the script is always the first parameter in the sys.argv list, namely sys.argv[0].

C++ Learning Route and Future Employment Trend

https://qaofficial.com/post/2019/04/18/72978-c-learning-route-and-future-employment-trend.html 2019-04-18
It has been four years since learning C++, but it has never been deeply studied. Some people say that C++ is an object-oriented programming language, but what I want to say here is that C++ is a multi-generic programming language, which can be process-oriented. For example, we can write C code in C++, which is no problem. It is also an object-oriented language, which has object-oriented characteristics and can simulate various things in the real world.

C++ Programmer Development Direction

https://qaofficial.com/post/2019/04/18/72983-c-programmer-development-direction.html 2019-04-18
1. C++ Server Programmer (Streaming Media Background, Game Background, High Performance Server Background) 1. Proficient in C++, STL, Linux, etc., familiar with design patterns; 2. Proficient in a scripting language (Lua, Python, Perl, etc.); 3. Have a certain understanding of multi-thread environment programming, can independently complete the development, maintenance and optimization of server-side modules; 4. Proficient in MySQL database development and maintenance, performance optimization; 1. Proficient in C++ programming, at least 3 years experience in server development;

CNN Realize Character Recognition in Camera

https://qaofficial.com/post/2019/04/18/73035-cnn-realize-character-recognition-in-camera.html 2019-04-18
Before this, I have already said about the simple process of tensorflow. I will annotate the intermediate application functions in the example (see the function explanation in tensorflow for more details).Using cnn to realize character recognition in video, I wanted to talk about the principle of cnn first, but based on time and other people, I will say more here in great detail and give a direct example. 1 Face

CVPR 2018 Conference Resource Collection

https://qaofficial.com/post/2019/04/18/73011-cvpr-2018-conference-resource-collection.html 2019-04-18
1. Links to some visual conference proceedings Computer Vision Conference Proceedings Download: http://openaccess.thecvf.com/menu.py CVPR 2018, Salt Lake City Utah [Main Conference] [Workshops]   ICCV 2017, Venice Italy [Main Conference] [Workshops]   CVPR 2017, Honolulu Hawaii [Main Conference] [Workshops]   CVPR 2016, Las Vegas Nevada [Main Conference] [Workshops]   ICCV 2015, Santiago Chile [Main Conference] [Workshops]   CVPR 2015, Boston Massachusetts [Main Conference] [Workshops]   CVPR 2014, Columbus Ohio [Main Conference] [Workshops]   ICCV 2013, Sydney Australia [Main Conference] [Workshops]   CVPR 2013, Portland Oregon [Main Conference] [Workshops]       2, CVPR 2018 paper address

LDA and Naive Bayesian Combination-Image Classification

https://qaofficial.com/post/2019/04/18/73086-lda-and-naive-bayesian-combination-image-classification.html 2019-04-18
The original space is converted into feature space by LDA method, and then classified by Naive Bayesian method. 1) convert into feature space by LDA method, and use Deng Cai, Xiaofei He, Yuxiao Hu, Jiawei Han, and Thomas Huang,% "learning a spatially smooth subspace for face recognition", code LDA.m transformation in cvpr' 2007 literature function: [eigvector, eigvalue] = lda (x, gnd, options) %eigvector represents a feature vector;

OpenCV Machine Learning (1): Bayesian Classifier Implements Code Analysis

https://qaofficial.com/post/2019/04/18/73075-opencv-machine-learning-1-bayesian-classifier-implements-code-analysis.html 2019-04-18
OpenCV's machine learning class is defined in the ml.hpp file. The basic class is CvStatModel, from which other classifiers are inherited. Study the CvNormalBayesClassifier classifier today. 1. Class Definition The following class definitions are available in ml.hpp: class CV_EXPORTS_W CvNormalBayesClassifier : public CvStatModel { public: CV_WRAP CvNormalBayesClassifier(); virtual ~CvNormalBayesClassifier(); CvNormalBayesClassifier( const CvMat* trainData, const CvMat* responses, const CvMat* varIdx=0, const CvMat* sampleIdx=0 ); virtual bool train( const CvMat* trainData, const

Scikit-Learn machine learning notes -- MNIST

https://qaofficial.com/post/2019/04/18/73076-scikit-learn-machine-learning-notes-mnist.html 2019-04-18
Scikit-Learn machine learning notes-MNIST reference document: handson-ml import numpy as np import matplotlib.pyplot as plt from sklearn.linear_model import SGDClassifier # 加载MNIST数据集 def load_dataset(): from sklearn.datasets import fetch_mldata mnist = fetch_mldata('MNIST original', data_home='dataset') X, y = mnist['data'], mnist['target'] X_train, X_test, y_train, y_test = X[:60000], X[60000:], y[:60000], y[60000:] shuffle_index = np.random.permutation(60000) X_train, y_train = X_train[shuffle_index], y_train[shuffle_index] print('load mnist successfully\n', 'X_train shape is: ',