Tag：Category
-
Java inner class
Access rules for inner classes: 1. Internal classes can directly access members in external classes, including private. The reason why you can directly access members in an external class is that the internal class holds a reference to an external class. Format: external class. This 2. To access internal classes, external classes must establish internal […]
-
Yarn related usage
1. Upgrade yarn version Yarn version upgrade You can upgrade yarn by running the following command: 1: Make sure homebrew is installed $ brew upgrade yarn 2: Download the official script to upgrade $ curl -o- -L https://yarnpkg.com/install.sh | bash 2. Basic usage Initialize a new project yarn init Add dependency package yarn add [package] […]
-
Python machine learning perceptron
Recently I am reading books related to machine learning. By the way, I will write out the parts I read every day and share them with you. Let’s learn, discuss and make progress together! As the first blog of machine learning, I’m going to start with the perceptron and then update other content slowly. Before […]
-
Data preparation of fast RCNN
Data presentation Pascal VOC data set is a common data set in target detection and segmentation, which contains 20 categories. The directory structure of the VOC dataset is as follows. Annotations: save the annotation information of the picture in the form of XML file. ImageSets： a. The subfolder action stores data sets of actions of […]
-
Demo: build streaming application based on Flink SQL
All the practical exercises in this article will be executed on Flink SQL cli, only involving SQL plain text, without a line of Java / Scala code and IDE installation. Last Thursday, we shared Demo: building streaming applications based on Flink SQL live in the Chinese community of Flink nail group. The live content tends […]
-
Python machine learning — Logical Regression
We know that the perceptron algorithm can’t do anything for the data which can’t be completely linearly segmented. In this paper, we will introduce another very effective binary classification model – logical regression. It is widely used in classification tasks Logical regression is a classification model. Before implementation, we will introduce several concepts: Odds ratio: […]
-
Angular 2 Component Inheritance
Angular 2.3The version introduces the function of component inheritance, which is very powerful and can greatly increase the reusability of our components. Component Inheritance Component inheritance involves the following: Metadata: such [email protected]()、@Output()、@ContentChild/Children、@ViewChild/ChildrenAnd so on. The metadata defined in the derived class will overwrite any previous metadata in the inheritance chain, otherwise the base class metadata […]
-
Reading notes “machine learning in all aspects”
Feature Engineering Extract the features from the original data, and use the extracted results as the input of the algorithm and model. Normalize the numerical type characteristics: What’s this: Unify all data features into a roughly same numerical range, eliminate the dimensional influence between data features, and make different features comparable Common methods: The results […]
-
Let the machine read the video: uncover the multimodal AI algorithm behind the billion level Taobao video
background With the popularization of 4G and the launch of 5g, people pay more and more attention to the demand of content consumption. The 2019 Internet trend report points out that under the background of the overall slowdown of the mobile Internet industry, the short video industry has sprung up as a “black hole in […]
-
Machine learning support vector machine (SVM)
Catalog Support vector machine (SVM) 1. Fundamentals 2. Soft septum 3. Kernel function 4. Sklearn implements SVM 5. SVM multi classification 4.1 multi classification principle 4.2 SVM multi classification based on sklearn PrefaceRefer to machine learning. I don’t understand the dual problem…. I’m just a code Porter Machine learning column： Machine learning – linear regression […]
-
Common indicators of semantic segmentation
There are three most commonly used indicators in semantic segmentation. To facilitate interpretation, the confusion matrix needs to be introduced first, as follows: Confusion matrix True value Positive Negative predicted value Positive True positive (TP) False positive (FP) Negative False negative (FN) True negative (TN) First of all, it is assumed that […]