Announcing new ways to customize models with SageMaker AI, without managing GPUs
Amazon SageMaker AI 无服务器模型定制新增训练会话功能
要点导读
- 无服务器模型定制新增训练会话
- 用户可自定义后训练循环
- 平台托管 GPU 与分布式执行
- 支持故障恢复和检查点存储
原文摘要 · 中英对照
Amazon SageMaker AI serverless model customization now includes training sessions, generally available today. You write and coordinate your own post-training loop (custom loss functions, reward signals, or multi-turn reinforcement learning) while SageMaker AI handles GPU allocation, distributed execution, fault recovery, and checkpoint storage.
Amazon SageMaker AI 无服务器模型定制现包含训练会话,今日正式可用。您编写和协调自己的后训练循环(自定义损失函数、奖励信号或多轮强化学习),而 SageMaker AI 处理 GPU 分配、分布式执行、故障恢复和检查点存储。
阅读原文
原文来源:AWS 机器学习博客(aws.amazon.com)。本页摘要与要点整理自公开来源,版权归原作者所有,完整内容请阅读原文。