My main business is desktop development , Occasionally you need to do data and algorithms . Recently in use python Deal with some work , It's in full swing , All of a sudden python It can't be used , After checking the records, it turns out that IT The administrator forbids me to use python.
I don't know why IT The administrator will keep an eye on python, In a word, first submit to understand the requirements of the envelope . Waiting for IT I don't want to interrupt my work while I'm unsealed , I have to find a way to run it myself python. I know there are a lot of online runs python Environment , And Microsoft itself has Codespaces and Remote Development Extension Pack Things like that , But this time I'm going to try Azure Of Machine learning studio Run in python, By the way Azure Introduction article .
Azure The organizational structure of resources in , It has four levels ： Management Group 、 subscribe 、 Resource groups and resources .
Use Azure The premise is to have a Azure subscribe , If it's just trial and learning, there are several ways . Sign in Auzre Gateway after , Can find 200 Free trial in US dollars and Student rights The page of ：
You can also find any one docs.microsoft.com/zh-cn/learn A tutorial for , Open one Azure Sandbox . For example, you can This article Try to open a sandbox in the tutorial .Microsoft Learn Sandbox （ Sometimes called “Azure Sandbox ”） Can be used to pass Microsoft Learn Content exploration Azure Free environment for . A sandbox can only survive 4 Hours , You can create multiple sandboxes every day .
Select the subscription for the resource group you want to create , Enter the name of the resource group , Select area , And then you can click “ see + establish ” Button to create a resource group .
Now? , The subscription and resource group have been created , The next step is to create specific resources . Get into All services -> Mechanical learning page , Create a new machine learning workspace .
Choose your subscription on this page 、 Resource group 、 region , Just type in a name , And then click “ Review + establish " Button to complete the creation of machine learning workspace .
Wait a moment , Page prompt resources “ Deployment completed ”, Click on ” Go to resources “ Enter the newly created machine learning resource .
On the machine learning page , Click on ” Start the workspace “, Go to the machine learning studio page .
stay Azure The job of machine learning is , I can use Notebooks function python, That's my goal . There are other machine learning functions , That's not what I care about right now , Try again if you have a chance . But now it's time to run python It's one step away , I need another one Calculation example Used to run the python Code . Select... From the menu on the left “ Calculation “, And then in ” Calculation example “ Page hits ” establish “：
Choose a virtual machine type that you can afford , Enter the name and click ” establish “ You can complete the creation of a calculation instance .Azure Most of the commonly used tools have been pre installed in the preset machine learning computing examples , Basically out of the box .
Last , Select... From the menu on the left ”Notebooks“, Upload all kinds of documents and codes , open python Code runs , My goal has been achieved .
Last, last , Because I just want to run it temporarily python, Remember to close the calculation instance after using it . Even if the calculation instance is closed Azure There will still be some fees , Because all kinds of resources created can be regarded as renting hard disk space .
stay Cost management + charging You can view your own Azure Amount and cost and forecast , It's just running python It doesn't cost much ：
stay Azure In the course of using , All the exports on the page are links to tips and learning resources , for example Azure Machine learning documents . If you're not like me, just use it python, These resources should help you . You can also be in Learn Microsoft Docs I found some good courses in , Like this one ： Create machine learning models .
Participation of this paper Tencent cloud media sharing plan , You are welcome to join us , share .