Tag：Deep learning
-
Time：2019-11-22
By logm This article was originally published at https://segmentfault.com/u/logm/articles and is not allowed to be reproduced~ 1. Preface Tensorflow Lite is the version of tensorflow mobile. There are many blogs on the Internet about how to add custom OP for tensorflow. I will not introduce them. Because tensorflow Lite is relatively small, there are few […]
-
Time：2019-11-21
Catalog Five step life cycle of neural network model in keras Apply deep learning in Python mini course Binary classification course of keras deep learning library How to build multilayer perceptron neural network model with keras How to check the deep learning model in keras 10 command line scripts for Amazon Web Services deep learning […]
-
Time：2019-11-20
Absrtact: This paper is an in-depth learning entry project – digital handwriting recognition through keras. The whole process is introduced in detail, which is suitable for beginners to practice. For image classification task, convolutional neural network (CNN) is the best network structure, there is no one. CNN is widely used in face recognition, automatic driving, […]
-
Time：2019-11-18
Abstract: want to learn neural network? I don’t know how these things work. introduceWith the development of neural network, many tasks that were considered difficult to accomplish in the past have been completed, such as image recognition, speech recognition, and it has become easier to find deeper relationships in the data set. Therefore, we should […]
-
Time：2019-11-18
This article starts with vivo Internet technology WeChat public number https://mp.weixin.qq.com/s/wLMvJPXXaND9xq-XMwY2Mg.By eike dehlingTranslator: Yang Zhentao This article was released by eike dehling, a software and data engineer from textkernel, on October 23, 2018, on his pulse with LinkedIn, and has been authorized for translation.Original English link: https://www.linkedin.com/pulse/searching-deep-learning-ike-dehling/ Catalog: I. fast nearest neighborsII. Elasticsearch plug-inIII. integrationFour. […]
-
Time：2019-11-17
Abstract: want to learn deep learning? Can I count nothing in my heart? Special note: This article is to solve those who are totally unfamiliar with deep learning and are planning to enter this field. The aim is to help them think critically about the complexity of the field and to help them distinguish between […]
-
Time：2019-11-17
background For images, common tasks are: Image classification: extract category features, such as vgg19 network Target detection: extract category and location features, such as: Yolo network Semantic segmentation (instance segmentation): extract category, location feature,For each pixelFor example, deeplab network We know that when we do image classification, we usually add full connection layer + softmax […]
-
Time：2019-11-16
Overview of deep learning O restricted Boltzmann machine and deep belief network o Dropout O tips for dealing with imbalances O smote: synthesis of a few oversampling techniques O cost sensitive learning in neural networks Overview of deep learning Before 2006, the feedforward neural network of training depth supervision always failed, the main reason is […]
-
Time：2019-11-16
Abstract:How to define next generation storage standards? According to Alex Chen, senior director of Alibaba cloud intelligent storage products, its essence must have the following four characteristics: security and stability, optimization, seamless cloud access, and intelligence. The digital economy “rides the cloud”. Ten years ago, Alibaba cloud began to develop its own cloud computing operating […]
-
Time：2019-11-15
Alpha dogs have made deep learning the pinnacle of the data science world. Deep learning has become one of the most popular topics, but for most people, it is a strange and mysterious subject. Many people think that deep learning includes a lot of mathematical and statistical knowledge. This paper lists 12 common deep learning […]
-
Time：2019-11-15
origin Recently, I am learning neural network with my teacher. In order to understand the black box more deeply, I intend to implement it in C / C + +. Today, I have been busy for a long time, and finally realized a BP neural network. Later, I will successively realize CNN neural network and […]
-
Time：2019-11-14
Someone said it was an interview question!I didn’t say that, but it makes sense—Lu Xun ▍ what is an activation function In neurons, input is summed by weighting, and then a function is applied, which is the activation functionActivation Function。 The location of the activation function in the neural network is shown as follows: ▍ […]
