Custom translation service call guide

1. Custom model call API

1.1 Input parameters

1.2 Output parameters

Example

Input:

{
    "ModelId": "532",
    "Content": "hello",
    "ModelVersion": "V1" // You can specify the model version number, such as V1 or V2. If you do not specify this parameter, the latest version is used by default.
}

Outputs:

{"Data":["hello"],"RequestId":"fweo1j3931jd","Code":0}

Example of calling a custom model using an SDK

package com.alibaba.nlp.automl.modelcenter.service;
import com.aliyuncs.CommonRequest;
import com.aliyuncs.CommonResponse;
import com.aliyuncs.DefaultAcsClient;
import com.aliyuncs.IAcsClient;
import com.aliyuncs.http.MethodType;
import com.aliyuncs.profile.DefaultProfile;
public class PredictDemo {
    private void predictModel() {
        EnvironmentVariableCredentialsProvider credentialsProvider = CredentialsProviderFactory.newEnvironmentVariableCredentialsProvider();
      // Read the access credential from the environment variable.
        String regionId = "cn-hangzhou";
        try {
            // Create and initialize a DefaultAcsClient instance.
            DefaultProfile profile = DefaultProfile.getProfile("cn-hangzhou", credentialsProvider); 
            IAcsClient client = new DefaultAcsClient(profile);
            // Create an API request and set the parameters.
            CommonRequest request = new CommonRequest();
            request.setDomain("automl.cn-hangzhou.aliyuncs.com");
            request.setVersion("2019-07-01");
            request.setAction("PredictMTModel");
            request.setMethod(MethodType.POST);
            request.putQueryParameter("ModelId", "647");
            request.putQueryParameter("ModelVersion", "V1");
            request.putBodyParameter("Content", "hello");
            CommonResponse response = client.getCommonResponse(request);
            System.out.println(response);
        } catch (Exception e) {
          e.printStackTrace();
        }
    }
}

2. Intervention API

• Request URL: [http|https]://automl.cn-hangzhou.aliyuncs.com/api/automl/addMtIntervenePackage.

• You can call the AddMtIntervenePackage API operation over RPC to obtain results from the machine translation self-learning model.

2.1 Input parameters

2.2 Output parameters

Input:

{
        "PackageName": "test",
        "ProjectId": "1",
        "SourceLanguage": "zh",
        "TargetLanguage": "en",    
}

Outputs:

{
          "Code": "200",
          "RequestId": "aare83jnn9wj3",
          "Message": "",
          "PackageId": "1"
}

3. Add intervention word API

• Request URL: [http|https]://automl.cn-hangzhou.aliyuncs.com/api/automl/addMTInterveneWord.

• You can call the AddMTInterveneWord API operation over RPC to obtain results from the machine translation self-learning model.

3.1 Input parameters

3.2 Output parameters

Input:

{
        "PackageName": "test",
        "ProjectId": "1"
}

Outputs:

{
        "Code": "200",
        "RequestId": "aare83jnn9wj3",
        "Message": "",
        "PackageId": "1"
}

4. Bind intervention term package to model API

• Request URL: [http|https]://automl.cn-hangzhou.aliyuncs.com/api/automl/bindIntervenePackageAndModel.

• You can call the BindIntervenePackageAndModel API operation over RPC to obtain results from the machine translation self-learning model.

4.1 Input parameters

4.2 Output parameters

Input:

{
        "PackageId": "1",
        "ProjectId": "1",
        "ModelId": "1",
        "ModelVersion": "V1"
}

Outputs:

{
        "Code": "200",
        "RequestId": "aare83jnn9wj3",
        "Message": "",
        "Success": "true"
}

5. Error codes