以下是一种部署作为服务创建的SageMaker终点的解决方法，包含了代码示例：

1. 创建一个SageMaker模型：

import sagemaker
from sagemaker import get_execution_role
from sagemaker.amazon.amazon_estimator import get_image_uri

role = get_execution_role()
container = get_image_uri(boto3.Session().region_name, 'xgboost')

s3_model_location = 's3://your-s3-bucket/model.tar.gz'
model = sagemaker.model.Model(model_data=s3_model_location,
                              image=container,
                              role=role,
                              sagemaker_session=sagemaker.Session())

2. 创建一个SageMaker终点配置：

endpoint_config_name = 'your-endpoint-config-name'
instance_type = 'ml.m4.xlarge'

endpoint_config = sagemaker.session.s3_endpoint_config(s3_model_location,
                                                      endpoint_config_name,
                                                      instance_type=instance_type)

3. 创建一个SageMaker终点：

endpoint_name = 'your-endpoint-name'

sagemaker_session.create_endpoint(endpoint_name=endpoint_name,
                                  config_name=endpoint_config_name)

4. 等待终点创建完成：

sagemaker_session.wait_for_endpoint(endpoint=endpoint_name)

5. 使用终点进行推理：

from sagemaker.predictor import RealTimePredictor
from sagemaker.serializers import CSVSerializer

predictor = RealTimePredictor(endpoint=endpoint_name,
                              sagemaker_session=sagemaker.Session(),
                              serializer=CSVSerializer())

data = '1,2,3,4'  # 输入数据
response = predictor.predict(data)
print(response)

这样就完成了部署作为服务创建的SageMaker终点，并使用终点进行推理的过程。