Programmer Think

#Effective switching between py2 and py3 from __future__ import division, print_function, unicode_literals import numpy as np import os preface When carrying out a complete machine learning project, the following steps need to be completed Understand the complete projectget dataDiscover visual data, discover rulesFind the appropriat ...

Today, let's look at an interesting problem in data preprocessing: how to deal with missing values. Before we discuss the problem, let's review some basic terms to help us understand why we need to focus on missing values. The content of this article is long. I suggest learning after collection. I like to praise and support it. catalogue Int ...

1. Introduction of BP algorithm 1.1 Background Enlightened by the neural network in the human brain, computer simulation of the neural network in the human brain is used to realize machine learning technology of artificial intelligence. The bp algorithm is the most successful neural network algorithm to date. The bp algorithm can be used ...

Gated Channel Transformation for Visual Recognition (GCT) This article is a reproduction of an Attention article invested by Baidu & Syndney in CVPR2020. It proposes a general and lightweight change unit, which combines normalization method and Attention mechanism, which is easy to analyze the relationship between channels (competition or ...

GOAL: to train a model on a variety of learning tasks, such that it can solve new learning tasks using only a small number of training samples. Introduction: starting from face recognition, if a company has 50 people, it needs to make a face recognition system. According to the traditional deep learning idea, the recognition results should b ...

Wang Hui, joined qunar.com in 2017. At present, he is responsible for anti crawler related risk control business, has a wide range of technical fields, and is continuously exploring the practice direction of intelligent risk control. I preface In Qunar intelligent risk control scenario, the risk control R & D team often applies som ...

Note: this blog is based on Python 3, matplotlib 3.5.1, numpy 1.22.0 preface Visual operation is widely used in data analysis and machine learning. It can help people observe outliers or required data conversion effects more intuitively; As a desktop drawing package of python, matplotlib can generate publishing quality charts; The work ...

Related articles 1, Python getting started environment configuration: Anaconda3 and pycharm installation tutorial 2, labelimg application: dataset Tags 3, Build a pytoch environment under Windows (CPU version) 4, Building a pytoch environment under Windows (super detailed) 1, YOLOV5 environment construction (official requirements: ...

Tianchi competition -- visual analysis of user emotion catalogue Tianchi competition -- visual analysis of user emotion preface 1, Read the data, check the basic situation and preprocess the data Import related Library Read data, basic analysis data Null value processing, data mapping Word segmentation analysis of comments 2, Word ...

Lesson 4.3 basic principle and manual implementation of gradient descent In the previous section, we have successfully constructed the loss function of logistic regression, but due to the particularity of the logistic regression model itself, we cannot adopt the calculation idea similar to SSE when constructing the loss function (at this time, ...