azure openai studio api key的进一步展开说明

### **Integrating Azure OpenAI into your application**

Azure OpenAI is a powerful service that allows developers to build applications that can understand natural human language. It provides access to pre-trained AI models and a suite of APIs and tools for customization. In this blog post, we will guide you through the process of integrating Azure OpenAI into your own application to summarize text.

### **Provisioning an Azure OpenAI resource**

Before you can start using Azure OpenAI models, you need to provision an Azure OpenAI resource in your Azure subscription. Here are the steps to do so:

1. Sign into the Azure portal and create a new Azure OpenAI resource by providing the necessary details including the subscription, resource group, region, name, and pricing tier.

2. Once the deployment is complete, navigate to the Keys and Endpoint page of the deployed Azure OpenAI resource and save the keys and endpoint to a text file for later use.

### **Deploying a model**

To use the Azure OpenAI API, you first need to deploy a model through Azure OpenAI Studio. Here’s how:

1. On the Overview page of your Azure OpenAI resource, click the Explore button to open Azure OpenAI Studio in a new browser tab.

2. In Azure OpenAI Studio, create a new deployment with the model “gpt⑶5-turbo” and the default model version. Give the deployment a name, such as “text-turbo”.

### **Setting up an application in Cloud Shell**

To demonstrate how to integrate with an Azure OpenAI model, we will use a command-line application that runs in Cloud Shell on Azure. Follow these steps to set it up:

1. Open a new browser tab and go to the Azure portal.

2. Click on the Cloud Shell button (>_), located at the top of the page to the right of the search box, to open the Cloud Shell pane at the bottom of the portal.

3. If prompted, choose the Bash shell type for Cloud Shell. If you’re asked to create storage for your Cloud Shell, follow the instructions to do so.

4. Once the Cloud Shell is ready, run the following command to download the sample application and save it to a folder called “azure-openai”:

“`shell
rm -r azure-openai -f
git clone https://github.com/MicrosoftLearning/mslearn-openai azure-openai
“`

5. Navigate to the lab files for this exercise:

“`shell
cd azure-openai/Labfiles/02-nlp-azure-openai
“`

### **Configuring your application**

To enable the use of your Azure OpenAI resource, you need to complete some key parts of the application. Here’s how:

1. Open the configuration file for your preferred language in the code editor. For C#, it’s “appsettings.json”, and for Python, it’s “.env”.

2. Update the configuration values to include the endpoint, key, and model name from the Azure OpenAI resource you created. Save the file.

3. Navigate to the folder for your preferred language and install the necessary packages. For C#, run the following command:

“`shell
cd CSharp
dotnet add package Azure.AI.OpenAI –prerelease
“`

For Python, run the following commands:

“`shell
cd Python
pip install python-dotenv
pip install openai
“`

4. Open the code file for your preferred language in the code editor and add the necessary libraries.

5. Add the code for building the request, which specifies the parameters for your model, such as the prompt and temperature.

### **Running your application**

Now that your application is configured, you can run it to send requests to your Azure OpenAI model. Here’s how:

1. In the Cloud Shell bash terminal, navigate to the folder for your preferred language.

2. Run the application using the appropriate command. For C#, use:

“`shell
dotnet run
“`

For Python, use:

“`shell
python test-openai-model.py
“`

3. Observe the summarization of the sample text file.

4. To experiment with different temperature values, navigate to your code file and change the temperature value. Run the application again and observe the output.

### **Further explanations and examples**

By integrating Azure OpenAI into your application, you can create chatbots, language models, and other applications that excel at understanding natural human language. The Azure OpenAI service provides access to pre-trained AI models and a range of APIs and tools for customization. Its models are optimized for different balances of capabilities and performance. In this exercise, we used the GPT⑶ model family’s 3.5 Turbo model series, which is highly capable for language understanding.

To use Azure OpenAI models, you need an Azure subscription that has been approved for access to the Azure OpenAI service. You can sign up for a free Azure subscription and request access to the Azure OpenAI service through the provided links.

Once you have an approved Azure OpenAI subscription, you must provision an Azure OpenAI resource in your Azure subscription. This resource allows you to access and deploy models for your application. You can choose the subscription, resource group, region, name, and pricing tier for the resource. After deployment, you can find the keys and endpoint for your resource on the Keys and Endpoint page.

To use the Azure OpenAI API, you need to deploy a model through Azure OpenAI Studio. In our example, we deployed the “gpt⑶5-turbo” model with the default version and named it “text-turbo”. Once the model is deployed, you can reference it in your application.

We set up an application in Cloud Shell to demonstrate the integration with an Azure OpenAI model. Cloud Shell provides an environment for running command-line applications on Azure. We downloaded a sample application and saved it to the “azure-openai” folder. Then, we navigated to the lab files for this exercise.

To configure the application, we updated the configuration values in the appsettings.json or .env file to include the keys, endpoint, and model name from the Azure OpenAI resource we created. We also installed the necessary packages for our preferred language and added the required libraries to the code file.

In our application code, we built the request for the Azure OpenAI model, specifying parameters such as the prompt and temperature. The prompt is the initial input given to the model, and the temperature determines the randomness of the model’s output.

To run the application, we used the appropriate command in the Cloud Shell terminal. We observed the summarization of the sample text file and experimented with different temperature values to see how it affected the output.

### **Conclusion**

Integrating Azure OpenAI into your application allows you to take advantage of powerful AI models that excel at understanding natural human language. With Azure OpenAI, you can create chatbots, language models, and other applications that offer intelligent responses and summarization capabilities. By following the steps outlined in this article, you can easily provision an Azure OpenAI resource, deploy a model, set up an application in Cloud Shell, configure the application, and run your code. This will enable you to leverage the capabilities of Azure OpenAI in your own applications, opening up a world of possibilities for understanding and interacting with natural language.

azure openai studio api key的常见问答Q&A

问题1：Azure OpenAI 入门教程的内容是甚么？

答案：Azure OpenAI 入门教程是一系列教程，旨在帮助用户了解怎样使用 Azure OpenAI 服务进行自然语言处理任务。教程内容包括：

• 无需代码实现基于私有数据的 GPT 模型。
• 使用 Azure OpenAI 创建一个聊天机器人的服务。
• 用 Python 调用 Azure OpenAi API。
• 申请微软 Azure OpenAI 账号。
• 打造自己的 ChatGPT。
• 使用 Azure OpenAI 服务的高级语言模型。

问题2：如何获得 Azure OpenAI API Key？

答案：要获得 Azure OpenAI API Key，你可以依照以下步骤操作：

1. 进入 Azure 门户。
2. 找到你的 Azure OpenAI 资源。
3. 在资源概览页面中，你将找到 API Key，它将用于访问 Azure OpenAI 服务。

问题3：怎样在 Azure OpenAI 中生成文本？

答案：要在 Azure OpenAI 中生成文本，你可以依照以下步骤进行操作：

1. 进入 Azure OpenAI Studio。
2. 在操场中找到聊天选项并点击。
3. 在助理设置中，你将看到一个新的配置项 “Add your data”。
4. 在此处输入你想要生成的文本。
5. 点击生成按钮，将文本发送到 Azure OpenAI 服务，并生成相应结果。

请注意以上行述仅为参考信息，实际操作可能因 Azure OpenAI 服务的更新而有所改变。建议参考官方文档以获得最新的指南和教程。