Modelos testados

As seções dobráveis a seguir fornecem informações sobre modelos de aprendizado de máquina que foram testados pela equipe do Amazon SageMaker Neo. Expanda a seção dobrável com base em sua estrutura para verificar se um modelo foi testado.

nota

Esta não é uma lista abrangente de modelos que podem ser compilados com o Neo.

Consulte Estruturas compatíveis os operadores suportados pelo SageMaker AI Neo para descobrir se você pode compilar seu modelo com SageMaker o Neo.

DarkNet

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows
AlexNet
ResNet50	X	X		X	X	X		X	X
YOLOv2				X	X	X		X	X
YOLOv2_minúsculo	X	X		X	X	X		X	X
YOLOv3_416				X	X	X		X	X
YOLOv3_minúsculo	X	X		X	X	X		X	X

MXNet

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows
AlexNet			X
Densenet121			X
DenseNet201	X	X	X	X	X	X		X	X
GoogLeNet	X	X		X	X	X		X	X
InceptionV3				X	X	X		X	X
MobileNet0,75	X	X		X	X	X			X
MobileNet1,0	X	X	X	X	X	X			X
MobileNetV2_0.5	X	X		X	X	X			X
MobileNetV2_1.0	X	X	X	X	X	X	X	X	X
MobileNetV3_Large	X	X	X	X	X	X	X	X	X
MobileNetV3_Small	X	X	X	X	X	X	X	X	X
ResNeSt50				X	X			X	X
ResNet18_v1	X	X	X	X	X	X			X
ResNet18_v2	X	X		X	X	X			X
ResNet50_v1	X	X	X	X	X	X		X	X
ResNet50_v2	X	X	X	X	X	X		X	X
ResNext101_32x4d
ResNext50_32x4d	X		X	X	X			X	X
SENet_154				X	X	X		X	X
SE_ 50_32x4d ResNext	X	X		X	X	X		X	X
SqueezeNet1,0	X	X	X	X	X	X			X
SqueezeNet1.1	X	X	X	X	X	X		X	X
VGG11	X	X	X	X	X			X	X
Xception	X	X	X	X	X	X		X	X
darknet53	X	X		X	X	X		X	X
resnet18_v1b_0.89	X	X		X	X	X			X
resnet50_v1d_0.11	X	X		X	X	X			X
resnet50_v1d_0.86	X	X	X	X	X	X		X	X
ssd_512_mobilenet1.0_coco	X		X	X	X	X		X	X
ssd_512_mobilenet1.0_voc	X		X	X	X	X		X	X
ssd_resnet50_v1	X		X	X	X			X	X
yolo3_darknet53_coco	X			X	X			X	X
yolo3_mobilenet1.0_coco	X	X		X	X	X		X	X
deeplab_resnet50			X

Keras

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows
densenet121	X	X	X	X	X	X		X	X
densenet201	X	X	X	X	X	X			X
inception_v3	X	X		X	X	X		X	X
mobilenet_v1	X	X	X	X	X	X		X	X
mobilenet_v2	X	X	X	X	X	X		X	X
resnet152_v1				X	X				X
resnet152_v2				X	X				X
resnet50_v1	X	X	X	X	X			X	X
resnet50_v2	X	X	X	X	X	X		X	X
vgg16			X	X	X			X	X

ONNX

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows
AlexNet			X
mobilenetv2-1.0	X	X	X	X	X	X		X	X
resnet18v1	X			X	X				X
resnet18v2	X			X	X				X
resnet50v1	X		X	X	X			X	X
resnet50v2	X		X	X	X			X	X
resnet152v1				X	X	X			X
resnet152v2				X	X	X			X
squeezenet1.1	X		X	X	X	X		X	X
vgg19			X						X

PyTorch (FP32)

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Ambarella CV25	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows
densenet121	X	X	X	X	X	X	X		X	X
inception_v3		X			X	X	X		X	X
resnet152					X	X	X			X
resnet18	X	X			X	X	X			X
resnet50	X	X	X	X	X	X			X	X
squeezenet1.0	X	X			X	X	X			X
squeezenet1.1	X	X	X	X	X	X	X		X	X
yolov4					X	X
yolov5				X	X	X
fasterrcnn_resnet50_fpn					X	X
maskrcnn_resnet50_fpn					X	X

TensorFlow

TensorFlow

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Ambarella CV25	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows
densenet201	X	X	X	X	X	X	X		X	X
inception_v3	X	X	X		X	X	X		X	X
mobilenet100_v1	X	X	X		X	X	X			X
mobilenet100_v2.0	X	X	X		X	X	X		X	X
mobilenet130_v2	X	X			X	X	X			X
mobilenet140_v2	X	X	X		X	X	X		X	X
resnet50_v1.5	X	X			X	X	X		X	X
resnet50_v2	X	X	X	X	X	X	X		X	X
squeezeNet	X	X	X	X	X	X	X		X	X
mask_rcnn_inception_resnet_v2					X
ssd_mobilenet_v2					X	X
faster_rcnn_resnet50_lowproposals					X
rfcn_resnet101					X

TensorFlow.Keras

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows
DenseNet121	X	X		X	X	X		X	X
DenseNet201	X	X		X	X	X			X
InceptionV3	X	X		X	X	X		X	X
MobileNet	X	X		X	X	X		X	X
MobileNetv2	X	X		X	X	X		X	X
NASNetGrande				X	X			X	X
NASNetMóvel	X	X		X	X	X		X	X
ResNet101				X	X	X			X
ResNet101V2				X	X	X			X
ResNet152				X	X				X
ResNet152v2				X	X				X
ResNet50	X	X		X	X			X	X
ResNet50V2	X	X		X	X	X		X	X
VGG16				X	X			X	X
Xception	X	X		X	X	X		X	X

TensorFlow-Leve

TensorFlow-Lite (FP32)

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows	i.MX 8M Plus
densenet_2018_04_27	X			X	X	X			X
inception_resnet_v2_2018_04_27				X	X	X			X
inception_v3_2018_04_27				X	X	X			X	X
inception_v4_2018_04_27				X	X	X			X	X
mnasnet_0.5_224_09_07_2018	X			X	X	X			X
mnasnet_1.0_224_09_07_2018	X			X	X	X			X
mnasnet_1.3_224_09_07_2018	X			X	X	X			X
mobilenet_v1_0.25_128	X			X	X	X			X	X
mobilenet_v1_0.25_224	X			X	X	X			X	X
mobilenet_v1_0.5_128	X			X	X	X			X	X
mobilenet_v1_0.5_224	X			X	X	X			X	X
mobilenet_v1_0.75_128	X			X	X	X			X	X
mobilenet_v1_0.75_224	X			X	X	X			X	X
mobilenet_v1_1.0_128	X			X	X	X			X	X
mobilenet_v1_1.0_192	X			X	X	X			X	X
mobilenet_v2_1.0_224	X			X	X	X			X	X
resnet_v2_101				X	X	X			X
squeezenet_2018_04_27	X			X	X	X			X

TensorFlow-Lite (INT8)

Modelos da	ARM V8	ARM Mali	Ambarella CV22	Nvidia	Panorama	TI TDA4 VM	Qualcomm 03 QCS6	X86_Linux	X86_Windows	i.MX 8M Plus
inception_v1							X			X
inception_v2							X			X
inception_v3	X					X	X		X	X
inception_v4_299	X					X	X		X	X
mobilenet_v1_0.25_128	X					X			X	X
mobilenet_v1_0.25_224	X					X			X	X
mobilenet_v1_0.5_128	X					X			X	X
mobilenet_v1_0.5_224	X					X			X	X
mobilenet_v1_0.75_128	X					X			X	X
mobilenet_v1_0.75_224	X					X	X		X	X
mobilenet_v1_1.0_128	X					X			X	X
mobilenet_v1_1.0_224	X					X	X		X	X
mobilenet_v2_1.0_224	X					X	X		X	X
deeplab-v3_513							X