Shulou(Shulou.com)06/02 Report--

This article mainly analyzes the relevant knowledge points of R language MLP neural network based on Keras and how to build the environment. The content is detailed and easy to understand, and the operation details are reasonable, which has a certain reference value. If you are interested, you might as well follow the editor and learn more about "R language MLP neural network based on Keras and how to build the environment".

Intro

R language is the first computer language I use, and it is also one of the mainstream data analysis languages at present. It is often compared with python. R language has a lot of package to choose from in EDA, graphics and machine learning. However, deep learning is overtaken by python due to the lack of learning library and appropriate framework. But with the release of Keras on R, the two languages are tied again.

This article uses Tensorflow and Keras library to recognize handwritten digits on R language. The code is partly derived from "Getting started with Deep Learning using Keras and TensorFlow in R". Author: NSS: https://www.analyticsvidhya.com/blog/2017/06/getting-started-with-deep-learning-using-keras-in-r/?spm=5176.100239.blogcont109827.13.QVuLG8. These codes may not be easy for beginners to understand, and the omission of the previous computer configuration has taken me through a lot of pitfalls. In this article, I will explain the background installation and the code in detail.

Set up the environment and configure the local computer

Computer model: MacBook Air

System: Windows 8.1Professional Edition 64-bit operating system

Processor: Intel ®Core ™i5-5250U CPU @ 1.60GHz

Install memory (RAM): 8.0 GB

Video card: HD 4000 core graphics card

Install TensorFlow and Keras

Please install Python3.5 or 3.6 first. Download address: https://www.python.org/getit/. Currently, the latest version is 3.7.1, but the tensorflow win8 version currently only supports python3.5 and 3.6. IOS or LINUX may support version 2.x. We use this Python to install tensorflow.

My installation version is Python3.6.7 64bit, and the installation shows that I need C Runtime Update (KB2999226), that is, I need to update or to install KB2999226 patch. If your computer does not have this patch, then go to the official website to find a suitable version of your computer to download the update. Note that KB2919355 needs to be installed before subsequent installation.

When installing, choose to install the associated pip and so on. Then type python in cmd to view the version. As a computer rookie, I didn't know what cmd was at first. In fact, it is the 'terminal' in ios system, which is called command prompt in Chinese in win system.

C:\ Users\ user > python

After that, you need to download Visual Clearing 2015 redistributable on the official website to pip install tensorflow. Download address: https://www.microsoft.com/en-us/download/details.aspx?id=53587.

Type in cmd after the installation is complete

C:\ Users\ user > pip install tensorflow

If it shows that the xxxxx cannot be accessed, add-- user

C:\ Users\ user > pip install-- user tensorflow

Similarly, continue to install keras

C:\ Users\ user > pip install-- user keras

Now that some installations are ready, you can check whether the installation is completed in the following code in cmd. If no error is displayed, then the installation is successful.

C:\ Users\ user > python

Import tensorflow

Import keras

Next, go to the R language section!

Install R and Rstudio

If you have used R before, please ignore this paragraph.

It is very easy to install R, download directly from the official website: https://mirrors.tuna.tsinghua.edu.cn/CRAN/

Then download Rstudio, which is equivalent to the open-hanging version of R language, the interface is very friendly compared to R, and there are many auxiliary features. Download address: https://www.rstudio.com/products/rstudio/download/

# Note that Rstudio is based on R language. You need to download and install R language before you can install it.

Deep Learning MLP based on R language installs Tensorflow and Keras in Rstudio

This part of the code is derived from: https://www.analyticsvidhya.com/blog/2017/06/getting-started-with-deep-learning-using-keras-in-r/?spm=5176.100239.blogcont109827.13.QVuLG8. I have added some comments for reference.

First install Keras and tensorflow in RStudio

Install.packages ("devtools") # install R development kit devtools::install_github ("rstudio/keras") # download keras from github to Rinstall.packages ("keras") # you can also download CRAN's Keras package directly # one of the above two installation methods can be library (keras) # load keras package install_tensorflow () # to establish a tensorflow environment for R language, default is CPU version install_tensorflow (gpu=TRUE) # if you want to customize the installation For example, using GPU, using this line of code to preprocess MNIST datasets

After configuring the environment, we began to build the neural network, using the dataset_mnist () dataset. The MNIST dataset consists of 60000 grayscale images of ten handwritten digits from 0 to 9, 28 x 28 pixels, while providing a test set of 10000 images.

First, we download the dataset and create variables for the test and training data. Where x is a 3D array of grayscale values (image, width, height), and y is an integer classification vector of the numbers 0 to 9.

# separate the training set and test set and create variables # this step is very common in deep learning and can effectively prevent data fraud and make your data and steps look clearer train_x

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.