Method 1: Use GUI-based Software for Inference (Postman)

1. Download Postman and install it, or install the Postman Chrome extension. Alternatively, use other software that can send POST requests. Postman 7.24.0 is recommended.
2. Open Postman. Figure 1 shows the Postman interface.
   Figure 1 Postman interface
   
3. Set parameters on Postman. The following uses image classification as an example.
   • Select a POST task and copy the API URL to the POST text box. To obtain the API URL of the real-time service, switch to the Usage Guides tab on the page providing details about the real-time service. On the Headers tab page, set Key to X-Auth-Token and Value to the obtained token.

     For details about how to obtain a token, see Obtaining a User Token. The real-time service APIs generated by ModelArts do not support tokens whose scope is domain. Therefore, you need to obtain the token whose scope is project.

     

     You can also use the AK and SK to encrypt API calling requests. For details, see SDK Reference > Session Authentication > AK/SK-based Authentication.

     Figure 2 Parameter settings
     
   • On the Body tab page, file input and text input are available.
     • File input

       Select form-data. Set KEY to the input parameter of the model, for example, images. Set VALUE to an image to be inferred (only one image can be inferred). See Figure 3.

       Figure 3 Setting parameters on the Body tab page
       
     • Text input

       Select raw and then JSON(application/json). Enter the request body in the text box below. An example request body is as follows:

       {
         "meta": {
           "uuid": "10eb0091-887f-4839-9929-cbc884f1e20e"
         },
         "data": {
           "req_data": [
             {
               "sepal_length": 3,
               "sepal_width": 1,
               "petal_length": 2.2,
               "petal_width": 4
             }
           ]
         }
       }

       meta can carry a universally unique identifier (UUID). When you call an API, the system provides a UUID. When the inference result is returned, the UUID is returned to trace the request. If you do not need this function, leave meta blank. data contains a req_data array for one or multiple pieces of input data. The parameters of each piece of data are determined by the model, for example, sepal_length and sepal_width in this example.

4. After setting the parameters, click Send to send the request. The result is displayed in the response.
   • Inference result using file input: Figure 4 shows an example. The field values in the return result vary with the model.
   • Inference result using text input: Figure 5 shows an example. The request body contains meta and data. If the request contains uuid, uuid will be returned in the response. Otherwise, uuid is left blank. data contains a resp_data array for the inference results of one or multiple pieces of input data. The parameters of each result are determined by the model, for example, sepal_length and predictresult in this example.
     Figure 4 File inference result
     
     Figure 5 Text inference result
     

Method 2: Run the cURL Command to Send an Inference Request

The command for sending inference requests can be input as a file or text.