美国东部(弗吉尼亚州北部)(us-east-1) 的 Docker 注册表路径和示例代码 - Amazon SageMaker ECR 路径
Amazon Web Services 文档中描述的 Amazon Web Services 服务或功能可能因区域而异。要查看适用于中国区域的差异,请参阅 中国的 Amazon Web Services 服务入门 (PDF)

美国东部(弗吉尼亚州北部)(us-east-1) 的 Docker 注册表路径和示例代码

以下主题列出了 Amazon SageMaker 在此区域中提供的各种算法和深度学习容器的参数。

AutoGluon(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='autogluon',region='us-east-1',image_scope='inference',version='0.4') # Output path '763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:0.4-cpu-py38'
注册表路径 版本 作业类型(映像范围)
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.7.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.7.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.6.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.6.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.6.1 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.6.1 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.5.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.5.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.4.3 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.4.3 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.4.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.4.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.4.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.4.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.3.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.3.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-training:<标签> 0.3.1 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/autogluon-inference:<标签> 0.3.1 推理

BlazingText(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='blazingtext',region='us-east-1') # Output path '811284229777.dkr.ecr.us-east-1.amazonaws.com/blazingtext:1'
注册表路径 版本 作业类型(映像范围)
811284229777.dkr.ecr.us-east-1.amazonaws.com/blazingtext:<标签> 1 推理、训练

Chainer (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='chainer',region='us-east-1',version='5.0.0',py_version='py3',image_scope='inference',instance_type='ml.c5.4xlarge') # Output path '520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-chainer:5.0.0-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-chainer:<标签> 5.0.0 推理、训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-chainer:<标签> 4.1.0 推理、训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-chainer:<标签> 4.0.0 推理、训练 CPU,GPU py2、py3

Clarify(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='clarify',region='us-east-1',version='1.0',image_scope='processing') # Output path '205585389593.dkr.ecr.us-east-1.amazonaws.com/sagemaker-clarify-processing:1.0'
注册表路径 版本 作业类型(映像范围)
205585389593.dkr.ecr.us-east-1.amazonaws.com/sagemaker-clarify-processing:<标签> 1.0 处理

DJL DeepSpeed(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='djl-deepspeed', region='us-west-2',py_version='py3',image_scope='inference') # Output path '763104351884.dkr.ecr.us-west-2.amazonaws.com/djl-inference:0.20.0-deepspeed0.7.5-cu116'
注册表路径 版本 作业类型(映像范围)
763104351884.dkr.ecr.us-east-1.amazonaws.com/djl-inference:0.22.1-deepspeed0.8.3-cu118-<标签> 0.22.1 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/djl-inference:0.21.0-deepspeed0.8.3-cu117-<标签> 0.21.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/djl-inference:0.20.0-deepspeed0.7.5-cu116-<标签> 0.20.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/djl-inference:0.19.0-deepspeed0.7.3-cu113-<标签> 0.19.0 推理

Data Wrangler(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='data-wrangler',region='us-east-1') # Output path '663277389841.dkr.ecr.us-east-1.amazonaws.com/sagemaker-data-wrangler-container:1.x'
注册表路径 版本 作业类型(映像范围)
663277389841.dkr.ecr.us-east-1.amazonaws.com/sagemaker-data-wrangler-container:<标签> 2.x 处理
663277389841.dkr.ecr.us-east-1.amazonaws.com/sagemaker-data-wrangler-container:<标签> 1.x 处理

调试器(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='debugger',region='us-east-1') # Output path '503895931360.dkr.ecr.us-east-1.amazonaws.com/sagemaker-debugger-rules:latest'
注册表路径 版本 作业类型(映像范围)
503895931360.dkr.ecr.us-east-1.amazonaws.com/sagemaker-debugger-rules:<标签> 最新 调试器

DeepAR 预测(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='forecasting-deepar',region='us-east-1') # Output path '522234722520.dkr.ecr.us-east-1.amazonaws.com/forecasting-deepar:1'
注册表路径 版本 作业类型(映像范围)
522234722520.dkr.ecr.us-east-1.amazonaws.com/forecasting-deepar:<标签> 1 推理、训练

因子分解机(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='factorization-machines',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/factorization-machines:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/factorization-machines:<标签> 1 推理、训练

Hugging Face(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='huggingface',region='us-east-1',version='4.4.2',image_scope='training',base_framework_version='tensorflow2.4.1') # Output path '763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:2.4.1-transformers4.4.2-gpu-py37'
注册表路径 版本 作业类型(映像范围)
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.26.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-inference:<标签> 4.26.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-inference:<标签> 4.26.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.17.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.17.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-inference:<标签> 4.17.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-inference:<标签> 4.17.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.12.3 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.12.3 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-inference:<标签> 4.12.3 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-inference:<标签> 4.12.3 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.11.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.11.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-inference:<标签> 4.11.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-inference:<标签> 4.11.0 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.10.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.10.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.10.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.10.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-inference:<标签> 4.10.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-inference:<标签> 4.10.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-inference:<标签> 4.10.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-inference:<标签> 4.10.2 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.6.1 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.6.1 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.6.1 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.6.1 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-inference:<标签> 4.6.1 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-inference:<标签> 4.6.1 推理
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.5.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.5.0 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:<标签> 4.4.2 训练
763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-tensorflow-training:<标签> 4.4.2 训练

IP 洞察(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='ipinsights',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/ipinsights:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/ipinsights:<标签> 1 推理、训练

图像分类(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='image-classification',region='us-east-1') # Output path '811284229777.dkr.ecr.us-east-1.amazonaws.com/image-classification:1'
注册表路径 版本 作业类型(映像范围)
811284229777.dkr.ecr.us-east-1.amazonaws.com/image-classification:<标签> 1 推理、训练

Inferentia MXNet (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='inferentia-mxnet',region='us-east-1',version='1.5.1',instance_type='ml.inf1.6xlarge') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-mxnet:1.5.1-inf-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-mxnet:<标签> 1.8 推理 inf py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-mxnet:<标签> 1.5.1 推理 inf py3

Inferentia PyTorch (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='inferentia-pytorch',region='us-east-1',version='1.9',py_version='py3') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-pytorch:1.9-inf-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-pytorch:<标签> 1.9 推理 inf py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-pytorch:<标签> 1.8 推理 inf py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-pytorch:<标签> 1.7 推理 inf py3

K-Means(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='kmeans',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/kmeans:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/kmeans:<标签> 1 推理、训练

KNN(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='knn',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/knn:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/knn:<标签> 1 推理、训练

LDA(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='lda',region='us-east-1') # Output path '766337827248.dkr.ecr.us-east-1.amazonaws.com/lda:1'
注册表路径 版本 作业类型(映像范围)
766337827248.dkr.ecr.us-east-1.amazonaws.com/lda:<标签> 1 推理、训练

线性学习器(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='linear-learner',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/linear-learner:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/linear-learner:<标签> 1 推理、训练

MXNet (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='mxnet',region='us-east-1',version='1.4.1',py_version='py3',image_scope='inference', instance_type='ml.c5.4xlarge') # Output path '763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference:1.4.1-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-training:<标签> 1.9.0 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference:<标签> 1.9.0 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-training:<标签> 1.8.0 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference:<标签> 1.8.0 推理 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-training:<标签> 1.7.0 训练 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference:<标签> 1.7.0 推理 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference-eia:<标签> 1.7.0 eia CPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-training:<标签> 1.6.0 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference:<标签> 1.6.0 推理 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference-eia:<标签> 1.5.1 eia CPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.4.1 训练 CPU,GPU py2
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-training:<标签> 1.4.1 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet-serving:<标签> 1.4.1 推理 CPU,GPU py2
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference:<标签> 1.4.1 推理 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/mxnet-inference-eia:<标签> 1.4.1 eia CPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.4.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet-serving:<标签> 1.4.0 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet-serving-eia:<标签> 1.4.0 eia CPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.3.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.3.0 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet-eia:<标签> 1.3.0 eia CPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.2.1 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.2.1 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.1.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.1.0 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.0.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 1.0.0 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 0.12.1 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-mxnet:<标签> 0.12.1 推理 CPU,GPU py2、py3

MXNet Coach (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='coach-mxnet',region='us-east-1',version='0.11',py_version='py3',image_scope='training',instance_type='ml.c5.4xlarge') # Output path '520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-mxnet:coach0.11-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-mxnet:coach0.11.0-<标签> 0.11.0 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-mxnet:coach0.11-<标签> 0.11 训练 CPU,GPU py3

模型监控器(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='model-monitor',region='us-east-1') # Output path '156813124566.dkr.ecr.us-east-1.amazonaws.com/sagemaker-model-monitor-analyzer'
注册表路径 版本 作业类型(映像范围)
156813124566.dkr.ecr.us-east-1.amazonaws.com/sagemaker-model-monitor-analyzer:<标签> 监控

NTM(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='ntm',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/ntm:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/ntm:<标签> 1 推理、训练

Neo Image Classification(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='image-classification-neo',region='us-east-1') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/image-classification-neo:latest'
注册表路径 版本 作业类型(映像范围)
785573368785.dkr.ecr.us-east-1.amazonaws.com/image-classification-neo:<标签> 最新 推理

Neo MXNet (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='neo-mxnet',region='us-east-1',version='1.8',py_version='py3',image_scope='inference', instance_type='ml.c5.4xlarge') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-mxnet:1.8-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-mxnet:<标签> 1.8 推理 CPU,GPU py3

Neo PyTorch (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='neo-pytorch',region='us-east-1',version='1.6',image_scope='inference',instance_type='ml.c5.4xlarge') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:1.6-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:<标签> 1.13 推理 CPU,GPU py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:<标签> 1.12 推理 CPU,GPU py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:<标签> 1.8 推理 CPU,GPU py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:<标签> 1.7 推理 CPU,GPU py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:<标签> 1.6 推理 CPU,GPU py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:<标签> 1.5 推理 CPU,GPU py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-pytorch:<标签> 1.4 推理 CPU,GPU py3

Neo Tensorflow (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='neo-tensorflow',region='us-east-1',version='1.15.3',py_version='py3',instance_type='ml.c5.4xlarge') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-tensorflow:1.15.3-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-tensorflow:<标签> 2.9.2 推理 CPU,GPU py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-inference-tensorflow:<标签> 1.15.3 推理 CPU,GPU py3

Neo XGBoost(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='xgboost-neo',region='us-east-1') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/xgboost-neo:latest'
注册表路径 版本 作业类型(映像范围)
785573368785.dkr.ecr.us-east-1.amazonaws.com/xgboost-neo:<标签> 最新 推理

对象检测(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='object-detection',region='us-east-1') # Output path '811284229777.dkr.ecr.us-east-1.amazonaws.com/object-detection:1'
注册表路径 版本 作业类型(映像范围)
811284229777.dkr.ecr.us-east-1.amazonaws.com/object-detection:<标签> 1 推理、训练

Object2Vec(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='object2vec',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/object2vec:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/object2vec:<标签> 1 推理、训练

PCA(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='pca',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/pca:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/pca:<标签> 1 推理、训练

PyTorch (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='pytorch',region='us-east-1',version='1.8.0',py_version='py3',image_scope='inference', instance_type='ml.c5.4xlarge') # Output path '763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:1.8.0-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 2.0.0 推理 CPU,GPU py310
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference-graviton:<标签> 2.0.0 inference_graviton CPU py310
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 2.0.0 训练 CPU,GPU py310
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.13.1 推理 CPU,GPU py39
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.13.1 训练 CPU,GPU py39
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.12.1 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference-graviton:<标签> 1.12.1 inference_graviton CPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.12.1 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.12.0 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.12.0 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.11.0 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.11.0 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.10.2 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.10.2 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.10.0 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.10.0 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.9.1 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.9.1 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.9.0 推理 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.9.0 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.8.1 推理 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.8.1 训练 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.8.0 推理 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.8.0 训练 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.7.1 推理 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.7.1 训练 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.6.0 推理 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.6.0 训练 CPU,GPU py3、py36
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference-eia:<标签> 1.5.1 eia CPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.5.0 推理 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.5.0 训练 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.4.0 推理 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.4.0 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference-eia:<标签> 1.3.1 eia CPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.3.1 推理 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.3.1 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-inference:<标签> 1.2.0 推理 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:<标签> 1.2.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-pytorch:<标签> 1.1.0 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-pytorch:<标签> 1.1.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-pytorch:<标签> 1.0.0 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-pytorch:<标签> 1.0.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-pytorch:<标签> 0.4.0 推理 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-pytorch:<标签> 0.4.0 训练 CPU,GPU py2、py3

PyTorch Neuron (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='pytorch-neuron',region='us-west-2', image_scope='inference') # Output path '763104351884.dkr.ecr.us-west-2.amazon.aws.com/pytorch-neuron:1.10.2-neuron-py37-sdk1.19.0-ubuntu18.04'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training-neuron:<标签> 1.11.0 训练 TRN py38

PyTorch Training Compiler (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='pytorch-training-compiler',region='us-west-2', version='py38') # Output path '763104351884.dkr.ecr.us-west-2.amazonaws.com/pytorch-trcomp-training:1.12.0-gpu-py38-cu113-ubuntu20.04-sagemaker'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-trcomp-training:<标签> 1.13.1 训练 GPU py39
763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-trcomp-training:<标签> 1.12.0 训练 GPU py38

Random Cut Forest(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='randomcutforest',region='us-east-1') # Output path '382416733822.dkr.ecr.us-east-1.amazonaws.com/randomcutforest:1'
注册表路径 版本 作业类型(映像范围)
382416733822.dkr.ecr.us-east-1.amazonaws.com/randomcutforest:<标签> 1 推理、训练

Ray PyTorch (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='ray-pytorch',region='us-east-1',version='0.8.5',instance_type='ml.c5.4xlarge') # Output path '462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-ray-container:ray-0.8.5-torch-cpu-py36'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-ray-container:ray-1.6.0-torch-<标签> 1.6.0 训练 CPU,GPU py36
462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-ray-container:ray-0.8.5-torch-<标签> 0.8.5 训练 CPU,GPU py36

Scikit-learn(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='sklearn',region='us-east-1',version='0.23-1',image_scope='inference') # Output path '683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:0.23-1-cpu-py3'
注册表路径 版本 程序包版本 作业类型(映像范围)
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 1.2-1 1.2.1 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 1.2-1 1.2.1 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 1.0-1 1.0.2 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 1.0-1 1.0.2 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 1.0-1 1.0.2 inference_graviton
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 0.23-1 0.23.2 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 0.23-1 0.23.2 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 0.20.0 0.20.0 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-scikit-learn:<标签> 0.20.0 0.20.0 训练

语义分割(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='semantic-segmentation',region='us-east-1') # Output path '811284229777.dkr.ecr.us-east-1.amazonaws.com/semantic-segmentation:1'
注册表路径 版本 作业类型(映像范围)
811284229777.dkr.ecr.us-east-1.amazonaws.com/semantic-segmentation:<标签> 1 推理、训练

Seq2Seq(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='seq2seq',region='us-east-1') # Output path '811284229777.dkr.ecr.us-east-1.amazonaws.com/seq2seq:1'
注册表路径 版本 作业类型(映像范围)
811284229777.dkr.ecr.us-east-1.amazonaws.com/seq2seq:<标签> 1 推理、训练

Spark(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='spark',region='us-east-1',version='3.0',image_scope='processing') # Output path '173754725891.dkr.ecr.us-east-1.amazonaws.com/sagemaker-spark-processing:3.0-cpu'
注册表路径 版本 作业类型(映像范围)
173754725891.dkr.ecr.us-east-1.amazonaws.com/sagemaker-spark-processing:<标签> 3.2 处理
173754725891.dkr.ecr.us-east-1.amazonaws.com/sagemaker-spark-processing:<标签> 3.1 处理
173754725891.dkr.ecr.us-east-1.amazonaws.com/sagemaker-spark-processing:<标签> 3.0 处理
173754725891.dkr.ecr.us-east-1.amazonaws.com/sagemaker-spark-processing:<标签> 2.4 处理

SparkML Serving(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='sparkml-serving',region='us-east-1',version='2.4') # Output path '683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-sparkml-serving:2.4'
注册表路径 版本 作业类型(映像范围)
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-sparkml-serving:<标签> 3.3 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-sparkml-serving:<标签> 2.4 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-sparkml-serving:<标签> 2.2 推理

Tensorflow (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='tensorflow',region='us-east-1',version='1.12.0',image_scope='inference',instance_type='ml.c5.4xlarge') # Output path '520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-serving:1.12.0-cpu'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.12.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.12.0 训练 CPU,GPU py310
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.11.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.11.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.11.0 训练 CPU,GPU py39
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.10.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.10.1 训练 CPU,GPU py39
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.10.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.9.3 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.9.2 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.9.2 训练 CPU,GPU py39
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference-graviton:<标签> 2.9.1 inference_graviton CPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.8.4 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.8.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.8.0 训练 CPU,GPU py39
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.7.1 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.7.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.6.3 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.6.3 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.6.2 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.6.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.6.0 训练 CPU,GPU py38
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.5.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.5.1 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.5.0 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.4.3 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.4.3 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.4.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.4.1 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.3.2 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.3.2 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.3.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.3.1 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference-eia:<标签> 2.3.0 eia CPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.3.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.3.0 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.2.2 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.2.2 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.2.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.2.1 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.2.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.2.0 训练 CPU,GPU py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.1.3 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.1.3 训练 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.1.2 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.1.2 训练 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.1.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.1.1 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.1.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.1.0 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.0.4 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.0.4 训练 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.0.3 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.0.3 训练 CPU,GPU py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.0.2 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.0.2 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.0.1 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.0.1 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference-eia:<标签> 2.0.0 eia CPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 2.0.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 2.0.0 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 1.15.5 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 1.15.5 训练 CPU,GPU py3、py36、py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 1.15.4 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 1.15.4 训练 CPU,GPU py3、py36、py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 1.15.3 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 1.15.3 训练 CPU,GPU py2、py3、py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 1.15.2 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 1.15.2 训练 CPU,GPU py2、py3、py37
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference-eia:<标签> 1.15.0 eia CPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 1.15.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 1.15.0 训练 CPU,GPU py2、py3
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference-eia:<标签> 1.14.0 eia CPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 1.14.0 推理 CPU,GPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 1.14.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-scriptmode:<标签> 1.13.1 训练 CPU,GPU py2
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-training:<标签> 1.13.1 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-serving-eia:<标签> 1.13.0 eia CPU -
763104351884.dkr.ecr.us-east-1.amazonaws.com/tensorflow-inference:<标签> 1.13.0 推理 CPU,GPU -
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-serving-eia:<标签> 1.12.0 eia CPU -
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-serving:<标签> 1.12.0 推理 CPU,GPU -
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-scriptmode:<标签> 1.12.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-serving-eia:<标签> 1.11.0 eia CPU -
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-serving:<标签> 1.11.0 推理 CPU,GPU -
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-scriptmode:<标签> 1.11.0 训练 CPU,GPU py2、py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow-eia:<标签> 1.10.0 eia CPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.10.0 推理 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.10.0 训练 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.9.0 推理 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.9.0 训练 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.8.0 推理 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.8.0 训练 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.7.0 推理 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.7.0 训练 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.6.0 推理 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.6.0 训练 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.5.0 推理 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.5.0 训练 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.4.1 推理 CPU,GPU py2
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-tensorflow:<标签> 1.4.1 训练 CPU,GPU py2

Tensorflow Coach (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='coach-tensorflow',region='us-east-1',version='1.0.0',image_scope='training',instance_type='ml.c5.4xlarge') # Output path '462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-coach-container:coach-1.0.0-tf-cpu-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-coach-container:coach-1.0.0-tf-<标签> 1.0.0 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:coach0.11.1-<标签> 0.11.1 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:coach0.11.0-<标签> 0.11.0 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:coach0.11-<标签> 0.11 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:coach0.10.1-<标签> 0.10.1 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:coach0.10-<标签> 0.10 训练 CPU,GPU py3

Tensorflow Inferentia (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='inferentia-tensorflow',region='us-east-1',version='1.15.0',instance_type='ml.inf1.6xlarge') # Output path '785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-tensorflow:1.15.0-inf-py3'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-tensorflow:<标签> 2.5.2 推理 inf py3
785573368785.dkr.ecr.us-east-1.amazonaws.com/sagemaker-neo-tensorflow:<标签> 1.15.0 推理 inf py3

Tensorflow Ray (DLC)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='ray-tensorflow',region='us-east-1',version='0.8.5',instance_type='ml.c5.4xlarge') # Output path '462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-ray-container:ray-0.8.5-tf-cpu-py36'
注册表路径 版本 作业类型(映像范围) 处理器类型 Python 版本
462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-ray-container:ray-1.6.0-tf-<标签> 1.6.0 训练 CPU,GPU py37
462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-ray-container:ray-0.8.5-tf-<标签> 0.8.5 训练 CPU,GPU py36
462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-ray-container:ray-0.8.2-tf-<标签> 0.8.2 训练 CPU,GPU py36
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:ray0.6.5-<标签> 0.6.5 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:ray0.6-<标签> 0.6 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:ray0.5.3-<标签> 0.5.3 训练 CPU,GPU py3
520713654638.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-tensorflow:ray0.5-<标签> 0.5 训练 CPU,GPU py3

VW(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='vw',region='us-east-1',version='8.7.0',image_scope='training') # Output path '462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-vw-container:vw-8.7.0-cpu'
注册表路径 版本 作业类型(映像范围)
462105765813.dkr.ecr.us-east-1.amazonaws.com/sagemaker-rl-vw-container:vw-8.7.0-<标签> 8.7.0 训练

XGBoost(算法)

用于检索注册表路径的 SageMaker Python SDK 示例。

from sagemaker import image_uris image_uris.retrieve(framework='xgboost',region='us-east-1',version='1.5-1') # Output path '683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:1.5-1'
注册表路径 版本 程序包版本 作业类型(映像范围)
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.7-1 1.7.4 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.7-1 1.7.4 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.5-1 1.5.2 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.5-1 1.5.2 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.5-1 1.5.2 inference_graviton
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.3-1 1.3.3 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.3-1 1.3.3 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.3-1 1.3.3 inference_graviton
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.2-2 1.2.0 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.2-2 1.2.0 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.2-1 1.2.0 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.2-1 1.2.0 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.0-1 1.0.0 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 1.0-1 1.0.0 训练
811284229777.dkr.ecr.us-east-1.amazonaws.com/xgboost:<标签> 1 0.72 推理
811284229777.dkr.ecr.us-east-1.amazonaws.com/xgboost:<标签> 1 0.72 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 0.90-2 0.90 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 0.90-2 0.90 训练
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 0.90-1 0.90 推理
683313688378.dkr.ecr.us-east-1.amazonaws.com/sagemaker-xgboost:<标签> 0.90-1 0.90 训练