Tag：gradient

Time：20211017
Introduction The development momentum of the indepth learning framework pytoch is amazing. Xiaobian also has a deep understanding of this. After flipping through the open source code of indepth learning on GitHub, he found that there are really many more people using pytoch. Therefore, Xiaobian is also going into pytoch recently, and writes articles to […]

Time：20211017
@TOC linear regression For the problem of classification, the data is discrete. For regression, the target value is continuous. The purpose of linear regression is to find a certain trend. In an ndimensional model. Its model can be used $$f(x)=w_1x_1+w_2x_2+…+w_dx_d+b$$The representation model is called a linear model. W is called the weight and B is […]

Time：20211014
The wechat “sweep” identification went online for a period of time. In the early stage, it mainly used the commodity map (shoes / bags / Beauty / clothing / household appliances / toys / books / food / Jewelry / furniture / others) as the medium to explore the valuable information in the wechat content […]

Time：20211011
This part introduces the failure cases of back propagation algorithm and the common methods of regularized neural network. Failure cases Many common situations will lead to the error of back propagation algorithm. Gradient disappearance The gradient of the lower layer (closer to the input) may become very small. In the depth network, the calculation of […]

Time：2021107
This article mainly talks about a very new feature of CSS, CSS @ property. Its emergence greatly enhances the ability of CSS! according toMDN — CSS Property, @ property CSS at rule is a part of CSS Houdini API. It allows developers to explicitly define their CSS custom properties, check property types, set default values, […]

Time：2021916
There are three gradients in fluent: LinearGradient: linear gradient RadialGradient: radial gradient Sweepgradient: Fan gradient Look at the original image. The following gradients are completed on the basis of this image. LinearGradient Add a linear gradient from top to bottom to a picture: ShaderMask( shaderCallback: (Rect bounds) { return LinearGradient( begin: Alignment.topCenter, end: Alignment.bottomCenter, colors: […]

Time：2021914
preface Sigmoid function (logistic function) is a very common activation function in neural networks. Let’s have a deep understanding of sigmoid function today. Function form Function image code implementation Code run: colab import matplotlib.pyplot as plt import numpy as np import math x = np.linspace(10, 10, 100) z = 1 / (1 + np.exp(x)) plt.title(“Sigmoid”) […]

Time：202191
Drconv is proposed in this paper, which combines the idea of local sharing and maintains translation invariance. It contains two key structures. From the experimental results, drconv meets the design expectations and has good performance improvement on multiple tasks Source: Xiaofei’s algorithm Engineering Notes official account Paper: dynamic region aware revolution Thesis address:https://arxiv.org/pdf/2003.12243.pdf Introduction At […]

Time：2021830
This paper proposes circle loss, which can not only deal with intra class optimization and inter class optimization separately, but also adjust the corresponding gradient according to different similarity values. In general, circle loss is more flexible, and the optimization goal is more clear. It has performed well in many experiments. Personally, I think it […]

Time：2021826
Hello, I’m Dongge. As one of the classical regression models in supervised learning, linear regression is a very good start for beginners. Considering the concept of comprehensiveness from a macro perspective, I think we may have been in contact in junior middle school. Y = ax, X is the independent variable, y is the dependent […]

Time：202189
By angel DasCompile VKSource: towards Data Science introduce Artificial neural networks (ANNs) are the advanced version of machine learning technology and the core of deep learning. Artificial neural networks involve the following concepts. Inputoutput layer, hidden layer, neurons under hidden layer, forward propagation and back propagation. Simply put, the input layer is a set of […]

Time：202187
CSS knowledge points Box model: W3C standard box model: width only includes content Ie box model: width only includes content + padding + border Clear floating method The purpose of clear floating is to solve the problem that the floating of child elements causes the high collapse of parent elements Pseudo class clear: both Common […]