简体繁体 English

多模型端点的无服务推理 - Amazon Sagemaker

[英]Serveless inference over multi-model endpoint - Amazon Sagemaker

原文 2023-01-10 10:21:29 1 1 amazon-sagemaker

I created a model on Sagemaker using the following two options.我使用以下两个选项在 Sagemaker 上创建了一个 model。 I also specified the URI for the custom container under ECR as well as the root path for the model archives.我还指定了 ECR 下自定义容器的 URI 以及 model 档案的根路径。

I am able to successfully created provisioned endpoint configuration however, in case of serverless, the following message showed up.我能够成功创建配置的端点配置，但是，在无服务器的情况下，会显示以下消息。 Does this mean that it is absolutely not possible on Sagemaker to have a serverless multimodel endpoint?这是否意味着 Sagemaker 上绝对不可能拥有无服务器多模型端点？

1 个解决方案

Does this mean that it is absolutely not possible on Sagemaker to have a serverless multimodel endpoint?这是否意味着 Sagemaker 上绝对不可能拥有无服务器多模型端点？ Basically with serverless you can deploy each model as a different endpoint and its cost effective as you pay only for usage.基本上使用无服务器，您可以将每个 model 部署为不同的端点，并且它具有成本效益，因为您只需为使用付费。 To answer your question technically you can't deploy multiple models on a serverless endpoint like you do with Multi model endpoints.要从技术上回答您的问题，您不能像使用 Multi model 端点那样在无服务器端点上部署多个模型。

贤者推理：如何加载model - Sagemaker inference : how to load model

为 PyTorch Model 调用 SageMaker 端点 - Invoking SageMaker Endpoint for PyTorch Model

使用 java sdk v2 调用 sagemaker 推理端点 - Call a sagemaker inference endpoint using the java sdk v2

调试和部署 sagemaker 端点的 featurizer（用于 imodel 推理的数据处理器） - debug and deploy featurizer (data processor for imodel inference) of sagemaker endpoint

AWS SageMaker Pipeline Model 端点部署失败 - AWS SageMaker Pipeline Model endpoint deployment failing

如何使用 Python SDK 调用 Amazon SageMaker 终端节点 - How do I invoke a Amazon SageMaker endpoint with the Python SDK

部署 TensorFlow 概率回归 model 作为 Sagemaker 端点 - Deploy TensorFlow probability regression model as Sagemaker endpoint

优化 sagemaker 上的批量转换推理 - Optimize batch transform inference on sagemaker

在 SageMaker 的处理过程中，在 source_dir 中使用 requirements.txt 重新打包推理 model 而无需安装它们 - Repack inference model with requirements.txt inside source_dir without installing them during the process in SageMaker

使用 Sagemaker 调用具有预训练自定义端点的调用超时 PyTorch model [推理] - Invocation timed out using Sagemaker to invoke endpoints with pretrained custom PyTorch model [Inference]

暂无

暂无

声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 贤者推理：如何加载model - Sagemaker inference : how to load model 为 PyTorch Model 调用 SageMaker 端点 - Invoking SageMaker Endpoint for PyTorch Model 使用 java sdk v2 调用 sagemaker 推理端点 - Call a sagemaker inference endpoint using the java sdk v2 调试和部署 sagemaker 端点的 featurizer（用于 imodel 推理的数据处理器） - debug and deploy featurizer (data processor for imodel inference) of sagemaker endpoint AWS SageMaker Pipeline Model 端点部署失败 - AWS SageMaker Pipeline Model endpoint deployment failing 如何使用 Python SDK 调用 Amazon SageMaker 终端节点 - How do I invoke a Amazon SageMaker endpoint with the Python SDK 部署 TensorFlow 概率回归 model 作为 Sagemaker 端点 - Deploy TensorFlow probability regression model as Sagemaker endpoint 优化 sagemaker 上的批量转换推理 - Optimize batch transform inference on sagemaker 在 SageMaker 的处理过程中，在 source_dir 中使用 requirements.txt 重新打包推理 model 而无需安装它们 - Repack inference model with requirements.txt inside source_dir without installing them during the process in SageMaker 使用 Sagemaker 调用具有预训练自定义端点的调用超时 PyTorch model [推理] - Invocation timed out using Sagemaker to invoke endpoints with pretrained custom PyTorch model [Inference]

相关标签

粤ICP备18138465号 © 2020-2024 STACKOOM.COM