Camera calibration and calibration

... 1. Calibration: First make a calibration board, OpenCV package has a black and white chessboard calibration board, direct A4 paper printing is OK. Then take a few photos with the camera that needs to be calibrated (Note: I have to take a few more shots from different angles, I used 22....


Deep learning - dropout understand

... 1.dropout solves the problem Deep neural network training is a very difficult thing, involving many factors, such as the local optimality caused by the non-convexity of the loss function Value, numerical stability during calculation, overfitting during training, etc. Among them, over...


Read all the images in the folder, display and output all the images in the specified folder

... into the folder, display and output to the specified folder. This includes obtaining the method of obtaining the number of files in a certain path, outputting it to the specified folder, and naming it according to the specified method. Use vs2015+opencv3.4.0 to achieve: #include <...


The combination of two ordered linked list sequences

... is known for two non-descending linked list sequences S1 and S2, and the design function constructs a new non-descending linked list S3 with S1 and S2 combined. Input format: Input is divided into two lines, each row gives a non-descending sequence of several positive integers, and −1...


Machine Learning - Dictionary Data Extraction

... 字典数据提取 from sklearn.feature_extraction import DictVectorizer Def dictverc(): """ Dictionary data extraction """ #描述化 Dict = DictVectorizer() #call fit_transform Data = dict.fit_transform([{'city': 'Beijing', 'temperature': 100}, {'city': 'Shanghai', 'temperature': 90}, {'city': 'Qingdao...


Combination feature (2) tfidf(word+article)+lsa

... """ Reduce the tfidf(word+article) feature to the lsa feature, save the result to the local, and save the result to the local """ From sklearn.decomposition import TruncatedSVD Import pickle Import time T_start = time.time() """======================================================== ==...


TensorFlow's Warning on AVX

... Downloads TensorFlow from Google. During the test, every time you run the code, you will output a lot of warning information: 2018-09-30 15:47:03.409806 : I tensorflow/core/platform/cpu_feature_guard.cc: 141] Your CPU supports instructions that this TensorFlow binary was not compiled to...


"Vertical Learning and Tensorflow" Reading Notes - Chapter 1: What is Machine Learning

... Machine Learning In terms of the types of learning, there are two types: 1. Unsupervised learning, that is, people After obtaining the trained vector data, try to find out the internal mining relationship in the case of without the label . 2. There is supervised learning, that is, each...


Machine Learning - Getting Started

... 前言(nonsense) I have been learning machine learning for some time. I have been taking notes with my mind map. I finally intend to share some knowledge in my learning process on the blog. This is my first public sharing. It is inevitable that I am somewhat unsatisfactory. I hope Don't...


Machine learning 100 days --- day06 NMF non-negative matrix factorization

... NMF, non-negative matrix factorization. This method can only be applied to data where each feature is non-negative. Basic idea: Decompose the large matrix into two small matrices, so that the two small matrices can be reduced to a large matrix as much as possible. The formula is: Vn∗m...