QA Official

【Scikit-learn] [Introduction] Six Functions of SCIKIT-LEARN

https://qaofficial.com/post/2019/05/04/24822-scikit-learn-introduction-six-functions-of-scikit-learn.html 2019-05-04
1. Introduction (1)SciPy, SciPy is an open source scientific computing toolkit based on Python. (2)Scikits, based on SciPy, has developed many branch versions for different application fields. They are collectively called Scikits, which means SciPy Toolkit (3) SCIKIT-LEARN is a branch version developed for machine learning based on SciPy. (4)Scikit-learn itself does not support deep learning. (5)Scikit-learn does not support GPU acceleration. Scikit-learn needs the support of other packages such as NumPy and SciPy, which is an open source framework developed by Python language for machine learning.

A Brief Introduction to the Lightweight Deep Learning Framework Keras

https://qaofficial.com/post/2019/05/04/24805-a-brief-introduction-to-the-lightweight-deep-learning-framework-keras.html 2019-05-04
徐海蛟教学 Keras是基于Theano的一个深度学习框架,它的设计参考了Torch,用Python语言编写,是一个高度模块化的神经网络库,支

ASP.NET is a common skill

https://qaofficial.com/post/2019/05/04/24866-asp.net-is-a-common-skill.html 2019-05-04
1. Database Access Performance Optimization Database Connection and Shutdown Access to database resources requires several operations, including creating connections, opening connections, and closing connections.These processes require multiple exchanges of information with the database to pass authentication, which consumes server resources.Connection Pool is provided in ASP.NET to improve the performance impact of opening and closing databases.The system places the user's database connection in the connection pool, takes it out when needed, withdraws the connection when closed, and waits for the next connection request.

Keras Learning Essay Based on Theano's Deep Learning Framework -10- Callback

https://qaofficial.com/post/2019/05/04/24862-keras-learning-essay-based-on-theanoamp#x27s-deep-learning-framework-10-callback.html 2019-05-04
Original Address: http://blog.csdn.net/niuwei22007/article/details/49229909 Callbacks (callback functions) are a set of functions used to be called at a specified stage during model training.You can view the internal information and statistical data of the model in the process of model training through the callback function.You can pass a callback function list to the fit () function, and then the related callback function can be called at the specified stage. 1, Callbacks base

Main Process of Installing and Using Keras on a New Ubuntu system

https://qaofficial.com/post/2019/05/04/24878-main-process-of-installing-and-using-keras-on-a-new-ubuntu-system.html 2019-05-04
First of all, you need to install the Ubuntu system, first make the installation USB stick, and follow the tutorial here: http://www.linuxidc.com/Linux/2016-04/130520.htm The first step is to install Universal USB Installer on Windows system. After searching, you can download it from this page provided by Baidu: http://rj.baidu.com/soft/detail/26320.html should be a non-toxic version. The second step is to make and install the USB flash drive. First, you need to download the

[reading notes] Keras Quick Start (1)

https://qaofficial.com/post/2019/05/04/24898-reading-notes-keras-quick-start-1.html 2019-05-04
"Keras Get Started Quickly" has a total of ten chapters. The part about deep learning can start directly from chapter 6. The recommendation in the installation book on Keras is to clone from Microsoft's server warehouse, install a git and add it to the environment variable, then clone from the git. As for the configuration part, Keras chooses one of CNTK,Tensorflow,Theano The ANO as the computing platform. In the book,

Fasttext text text classification

https://qaofficial.com/post/2019/05/03/24778-fasttext-text-text-classification.html 2019-05-03
1. Introduction 1, introduction fasttext is a word vector and text classification tool for facebook to open source. It will open source in 2016. The typical application scenario is "supervised text classification problem".Provides a simple and efficient text classification and Feature learning method with performance comparable to in-depth learning and faster speed. fastText combines the most successful concepts in natural language processing and machine learning.These include the use of word

Keras Introduction and Its Extensibility

https://qaofficial.com/post/2019/05/03/24804-keras-introduction-and-its-extensibility.html 2019-05-03
This article is reprinted from blog: https://blog.csdn.net/hewb14/article/details/53414068 Keras has good extensibility, on the one hand, because the interface is reserved at design time, and on the other hand, because the clear code structure allows you to have a lot of customization space.So here are a few examples of how to customize layers and various methods in Keras. 0、backend if you want to customize various layers and functions in Keras, you must use backend.

Keras Learning (3)-classification

https://qaofficial.com/post/2019/05/03/24628-keras-learning-3-classification.html 2019-05-03
This article mainly introduces the use of keras to build a neural network and classify handwritten numbers. code: import numpy as np from keras.datasets import mnist from keras.utils import np_utils from keras.models import Sequential from keras.layers import Dense, Activation from keras.optimizers import RMSprop # 使多次生成的随机数相同 np.random.seed(1337) # 下载数据集 # X_shape(60000 28x28),y shape(10000) (X_train, y_train), (X_test, y_test)

Keras Learning III: Realizing cifar10 Image Classification Model with CNN

https://qaofficial.com/post/2019/05/03/24636-keras-learning-iii-realizing-cifar10-image-classification-model-with-cnn.html 2019-05-03
Keras Learning III: Realizing cifar10 Image Classification Model with CNN 1 Introduction to Convolutional Neural Network Convolutional Neural Network, like fully connected neural networks, is formed by connecting multiple neural network layers.The difference is that CNN is generally composed of multiple convolution layers and pooling layers alternately connected to extract high-level features of input data and reduce the dimension of data.Finally, the extracted features are classified by neural network to