Las traducciones son generadas a través de traducción automática. En caso de conflicto entre la traducción y la version original de inglés, prevalecerá la version en inglés.
# Marcos, dispositivos, sistemas y arquitecturas compatibles
Amazon SageMaker Neo es compatible con los marcos de machine learning, dispositivos periféricos, sistemas operativos y arquitecturas de chips más comunes. Descubra si Neo es compatible con su estructura, dispositivo periférico, sistema operativo y arquitectura de chip seleccionando uno de los temas siguientes.
Encontrará una lista de los modelos que el equipo de Amazon SageMaker Neo ha probado en la sección [Modelos probados](neo-supported-edge-tested-models.md).
**nota**
Los dispositivos Ambarella requieren que se incluyan archivos adicionales en el archivo TAR comprimido antes de enviarlo para su compilación. Para obtener más información, consulte [Solución de errores de Ambarella](neo-troubleshooting-target-devices-ambarella.md).
Se requiere TIM-VX (libtim-vx.so) para i.MX 8M Plus. Para obtener información sobre cómo compilar TIM-VX, consulte el [repositorio de TIM-VX en GitHub](https://github.com/VeriSilicon/TIM-VX).
**Topics**
+ [Marcos admitidos](neo-supported-devices-edge-frameworks.md)
+ [Marcos, dispositivos, sistemas y arquitecturas compatibles](neo-supported-devices-edge-devices.md)
+ [Modelos probados](neo-supported-edge-tested-models.md)
# Marcos admitidos
Amazon SageMaker Neo es compatible con los siguientes marcos.
| Marcos | Versión de marco | Versión del modelo | Modelos | Formatos de modelo (empaquetados en \$1.tar.gz) | Kits de herramientas |
| --- | --- | --- | --- | --- | --- |
| MXNet | 1.8 | Compatible con 1.8 o versiones anteriores | Clasificación de imágenes, detección de objetos, segmentación semántica, estimación de poses, reconocimiento de actividades | Un archivo de símbolos (.json) y un archivo de parámetros (.params) | GluonCV v0.8.0 |
| ONNX | 1.7 | Compatible con 1.7 o versiones anteriores | Clasificación de imágenes, SVM | Un archivo de modelos (.onnx) | |
| Keras | 2.2 | Compatible con 2.2 o versiones anteriores | Clasificación de imágenes | Un archivo de definición de modelo (.h5) | |
| PyTorch | 1.7, 1.8 | Compatible con 1.7, 1.8 o versiones anteriores | Clasificación de imágenes, detección de objetos | Un archivo de definición de modelo (.pth) | |
| TensorFlow | 1.15, 2.4, 2.5 (solo para instancias ml.inf1.\$1) | Compatible con 1.15, 2.4, 2.5 (solo para instancias ml.inf1.\$1) o versiones anteriores | Clasificación de imágenes, detección de objetos | \$1En el caso de los modelos guardados, un archivo .pb o .pbtxt y un directorio de variables que contiene variables. \$1En el caso de modelos congelados, solo un archivo .pb o .pbtxt | |
| TensorFlow-LITE | 1.15 | Compatible con 1.15 o versiones anteriores | Clasificación de imágenes, detección de objetos | Un archivo de búfer plano de definición de modelo (.tflite) | |
| XGBoost | 1.3 | Compatible con 1.3 o versiones anteriores | Árboles de decisión | Un archivo de modelo de XGBoost (.model) en el que el número de nodos de un árbol es inferior a 2^31 | |
| DARKNET | | | Clasificación de imágenes, detección de objetos (el modelo Yolo no es compatible) | Un archivo de configuración (.cfg) y un archivo de pesos (.weights) | |
# Marcos, dispositivos, sistemas y arquitecturas compatibles
Amazon SageMaker Neo es compatible con los siguientes dispositivos, arquitecturas de chips y sistemas operativos.
## Dispositivos
Puede seleccionar un dispositivo mediante la lista desplegable de la [consola de Amazon SageMaker AI](https://console.aws.amazon.com/sagemaker) o especificando el `TargetDevice` en la configuración de salida de la API [https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateCompilationJob.html](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateCompilationJob.html).
Puede elegir uno de los siguientes dispositivos periféricos:
| Lista de dispositivos | Sistema en un chip (SoC) | Sistema operativo | Arquitectura | Acelerador | Opciones de compilación de ejemplo |
| --- | --- | --- | --- | --- | --- |
| aisage | Ninguna | Linux | ARM64 | Mali | Ninguna |
| amba\$1cv2 | CV2 | Arch Linux | ARM64 | cvflow | Ninguna |
| amba\$1cv22 | CV22 | Arch Linux | ARM64 | cvflow | Ninguna |
| amba\$1cv25 | CV25 | Arch Linux | ARM64 | cvflow | Ninguna |
| coreml | Ninguna | iOS, macOS | Ninguna | Ninguna | \$1"class\$1labels": "imagenet\$1labels\$11000.txt"\$1 |
| imx8qm | Mx8 de NXP | Linux | ARM64 | Ninguna | Ninguna |
| imx8mplus | i.MX 8M Plus | Linux | ARM64 | NPU | Ninguna |
| jacinto\$1tda4vm | TDA4VM | Linux | ARM | TDA4VM | Ninguna |
| jetson\$1nano | Ninguna | Linux | ARM64 | NVIDIA | \$1'gpu-code': 'sm\$153', 'trt-ver': '5.0.6', 'cuda-ver': '10.0'\$1Para `TensorFlow2`, `{'JETPACK_VERSION': '4.6', 'gpu_code': 'sm_72'}` |
| jetson\$1tx1 | Ninguna | Linux | ARM64 | NVIDIA | \$1'gpu-code': 'sm\$153', 'trt-ver': '6.0.1', 'cuda-ver': '10.0'\$1 |
| jetson\$1tx2 | Ninguna | Linux | ARM64 | NVIDIA | \$1'gpu-code': 'sm\$162', 'trt-ver': '6.0.1', 'cuda-ver': '10.0'\$1 |
| jetson\$1xavier | Ninguna | Linux | ARM64 | NVIDIA | \$1'gpu-code': 'sm\$172', 'trt-ver': '5.1.6', 'cuda-ver': '10.0'\$1 |
| qcs605 | Ninguna | Android | ARM64 | Mali | \$1'ANDROID\$1PLATFORM': 27\$1 |
| qcs603 | Ninguna | Android | ARM64 | Mali | \$1'ANDROID\$1PLATFORM': 27\$1 |
| rasp3b | ARM A56 | Linux | ARM\$1EABIHF | Ninguna | \$1'mattr': ['\$1neon']\$1 |
| rasp4b | ARM A72 | Ninguna | Ninguna | Ninguna | Ninguna |
| rk3288 | Ninguna | Linux | ARM\$1EABIHF | Mali | Ninguna |
| rk3399 | Ninguna | Linux | ARM64 | Mali | Ninguna |
| sbe\$1c | Ninguna | Linux | x86\$164 | Ninguna | \$1'mcpu': 'core-avx2'\$1 |
| sitara\$1am57x | AM57X | Linux | ARM64 | EVE y/o C66x DSP | Ninguna |
| x86\$1win32 | Ninguna | Windows 10 | X86\$132 | Ninguna | Ninguna |
| x86\$1win64 | Ninguna | Windows 10 | X86\$132 | Ninguna | Ninguna |
Para obtener más información sobre las opciones del compilador clave-valor JSON para cada dispositivo de destino, consulte el campo `CompilerOptions` del tipo de datos de la [ API `OutputConfig`](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_OutputConfig.html).
## Arquitecturas de sistemas y chips
Las siguientes tablas de consulta proporcionan información sobre los sistemas operativos y las arquitecturas disponibles para los trabajos de compilación de modelos Neo.
------
#### [ Linux ]
| Acelerador | X86\$164 | X86 | ARM64 | ARM\$1EABIHF | ARM\$1EABI |
| --- | --- | --- | --- | --- | --- |
| Sin acelerador (CPU) | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí |
| GPU Nvidia | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No |
| Intel\$1Graphics | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No |
| ARM Mali | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí |
------
#### [ Android ]
| Acelerador | X86\$164 | X86 | ARM64 | ARM\$1EABIHF | ARM\$1EABI |
| --- | --- | --- | --- | --- | --- |
| Sin acelerador (CPU) | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí |
| GPU Nvidia | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No |
| Intel\$1Graphics | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No |
| ARM Mali | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí |
------
#### [ Windows ]
| Acelerador | X86\$164 | X86 | ARM64 | ARM\$1EABIHF | ARM\$1EABI |
| --- | --- | --- | --- | --- | --- |
| Sin acelerador (CPU) | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/success_icon.svg) Sí | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No | ![\[alt text not found\]](http://docs.aws.amazon.com/es_es/sagemaker/latest/dg/images/negative_icon.svg) No |
------
# Modelos probados
Las siguientes secciones plegables proporcionan información sobre los modelos de machine learning que probó el equipo de Amazon SageMaker Neo. Amplíe la sección plegable en función de su estructura para comprobar si se ha probado un modelo.
**nota**
Esta no es una lista exhaustiva de los modelos que se pueden compilar con Neo.
Consulte [Marcos admitidos](neo-supported-devices-edge-frameworks.md) y [Operadores compatibles con SageMaker AI Neo](https://aws.amazon.com/releasenotes/sagemaker-neo-supported-frameworks-and-operators/) para averiguar si puede compilar el modelo con SageMaker Neo.
## DarkNet
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| AlexNet | | | | | | | | | |
| Resnet 50 | X | X | | X | X | X | | X | X |
| YoLoV2 | | | | X | X | X | | X | X |
| YoloV2\$1Tiny | X | X | | X | X | X | | X | X |
| Yolov3\$1416 | | | | X | X | X | | X | X |
| Yolov3\$1Tiny | X | X | | X | X | X | | X | X |
## MXNet
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| AlexNet | | | X | | | | | | |
| Densenet 121 | | | X | | | | | | |
| Densenet 201 | X | X | X | X | X | X | | X | X |
| Google Net | X | X | | X | X | X | | X | X |
| Inception v3 | | | | X | X | X | | X | X |
| MobileNet 0.75 | X | X | | X | X | X | | | X |
| MobileNet 1.0 | X | X | X | X | X | X | | | X |
| MobileNet V2\$10.5 | X | X | | X | X | X | | | X |
| MobileNet V2\$11.0 | X | X | X | X | X | X | X | X | X |
| MobileNet V3\$1Large | X | X | X | X | X | X | X | X | X |
| MobileNetV3\$1Small | X | X | X | X | X | X | X | X | X |
| Resest 50 | | | | X | X | | | X | X |
| ResNet18\$1v1 | X | X | X | X | X | X | | | X |
| ResNet18\$1v2 | X | X | | X | X | X | | | X |
| ResNet50\$1v1 | X | X | X | X | X | X | | X | X |
| ResNet50\$1v2 | X | X | X | X | X | X | | X | X |
| Resnext 101\$132x4D | | | | | | | | | |
| Resnext50\$132x4D | X | | X | X | X | | | X | X |
| Senet\$1154 | | | | X | X | X | | X | X |
| SE\$1Resnext50\$132x4D | X | X | | X | X | X | | X | X |
| SqueezeNet 1.0 | X | X | X | X | X | X | | | X |
| SqueezeNet 1.1 | 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\$1v1b\$10.89 | X | X | | X | X | X | | | X |
| resnet50\$1v1d\$10.11 | X | X | | X | X | X | | | X |
| resnet50\$1v1d\$10.86 | X | X | X | X | X | X | | X | X |
| ssd\$1512\$1mobilenet1.0\$1coco | X | | X | X | X | X | | X | X |
| ssd\$1512\$1mobilenet1.0\$1voc | X | | X | X | X | X | | X | X |
| ssd\$1resnet50\$1v1 | X | | X | X | X | | | X | X |
| yolo3\$1darknet53\$1coco | X | | | X | X | | | X | X |
| yolo3\$1mobilenet1.0\$1coco | X | X | | X | X | X | | X | X |
| deeplab\$1resnet50 | | | X | | | | | | |
## Keras
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| densenet 121 | X | X | X | X | X | X | | X | X |
| densenet 201 | X | X | X | X | X | X | | | X |
| inception\$1v3 | X | X | | X | X | X | | X | X |
| mobilenet\$1v1 | X | X | X | X | X | X | | X | X |
| mobilenet\$1v2 | X | X | X | X | X | X | | X | X |
| resnet152\$1v1 | | | | X | X | | | | X |
| resnet152\$1v2 | | | | X | X | | | | X |
| resnet50\$1v1 | X | X | X | X | X | | | X | X |
| resnet50\$1v2 | X | X | X | X | X | X | | X | X |
| vgg16 | | | X | X | X | | | X | X |
## ONNX
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| AlexNet | | | X | | | | | | |
| mobilenetv2-1.0 | X | X | X | X | X | X | | X | X |
| resnet 18 contra 1 | X | | | X | X | | | | X |
| resnet18 v2 | X | | | X | X | | | | X |
| resnet50 v1 | X | | X | X | X | | | X | X |
| resnet50 v2 | X | | X | X | X | | | X | X |
| resnet 152 v1 | | | | X | X | X | | | X |
| resnet 152 v2 | | | | X | X | X | | | X |
| squeezenet 1.1 | X | | X | X | X | X | | X | X |
| vgg19 | | | X | | | | | | X |
## PyTorch (FP32)
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Ambarella CV25 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| densenet 121 | X | X | X | X | X | X | X | | X | X |
| inception\$1v3 | | 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 |
| squeezenet 1.0 | X | X | | | X | X | X | | | X |
| squeezenet 1.1 | X | X | X | X | X | X | X | | X | X |
| yolov4 | | | | | X | X | | | | |
| yolov5 | | | | X | X | X | | | | |
| fasterrcnn\$1resnet50\$1fpn | | | | | X | X | | | | |
| maskrcnn\$1resnet50\$1fpn | | | | | X | X | | | | |
## TensorFlow
------
#### [ TensorFlow ]
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Ambarella CV25 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| densenet 201 | X | X | X | X | X | X | X | | X | X |
| inception\$1v3 | X | X | X | | X | X | X | | X | X |
| mobilenet100\$1v1 | X | X | X | | X | X | X | | | X |
| mobilenet100\$1v2.0 | X | X | X | | X | X | X | | X | X |
| mobilenet 130\$1v2 | X | X | | | X | X | X | | | X |
| mobilenet 140\$1v2 | X | X | X | | X | X | X | | X | X |
| resnet50\$1v1.5 | X | X | | | X | X | X | | X | X |
| resnet50\$1v2 | X | X | X | X | X | X | X | | X | X |
| SqueezeNet | X | X | X | X | X | X | X | | X | X |
| mask\$1rcnn\$1inception\$1resnet\$1v2 | | | | | X | | | | | |
| ssd\$1mobilenet\$1v2 | | | | | X | X | | | | |
| faster\$1rcnn\$1resnet50\$1lowproposals | | | | | X | | | | | |
| rfcn\$1resnet101 | | | | | X | | | | | |
------
#### [ TensorFlow.Keras ]
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| DenseNet 121 | X | X | | X | X | X | | X | X |
| Densenet 201 | X | X | | X | X | X | | | X |
| Inception v3 | X | X | | X | X | X | | X | X |
| MobileNet | X | X | | X | X | X | | X | X |
| MobileNet V2 | X | X | | X | X | X | | X | X |
| NASNetLarge | | | | X | X | | | X | X |
| NASNetMobile | X | X | | X | X | X | | X | X |
| ResNet101 | | | | X | X | X | | | X |
| ResNet 101 v2 | | | | X | X | X | | | X |
| ResNet 152 | | | | X | X | | | | X |
| Resnet 152 v2 | | | | X | X | | | | X |
| ResNet50 | X | X | | X | X | | | X | X |
| ResNet 50 v2 | X | X | | X | X | X | | X | X |
| VGG16 | | | | X | X | | | X | X |
| Xception | X | X | | X | X | X | | X | X |
------
## TensorFlow-LITE
------
#### [ TensorFlow-Lite (FP32) ]
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows | i.MX 8M Plus |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| densenet\$12018\$104\$127 | X | | | X | X | X | | | X | |
| inception\$1resnet\$1v2\$12018\$104\$127 | | | | X | X | X | | | X | |
| inception\$1v3\$12018\$104\$127 | | | | X | X | X | | | X | X |
| inception\$1v4\$12018\$104\$127 | | | | X | X | X | | | X | X |
| mnasnet\$10.5\$1224\$109\$107\$12018 | X | | | X | X | X | | | X | |
| mnasnet\$11.0\$1224\$109\$107\$12018 | X | | | X | X | X | | | X | |
| mnasnet\$11.3\$1224\$109\$107\$12018 | X | | | X | X | X | | | X | |
| mobilenet\$1v1\$10.25\$1128 | X | | | X | X | X | | | X | X |
| mobilenet\$1v1\$10.25\$1224 | X | | | X | X | X | | | X | X |
| mobilenet\$1v1\$10.5\$1128 | X | | | X | X | X | | | X | X |
| mobilenet\$1v1\$10.5\$1224 | X | | | X | X | X | | | X | X |
| mobilenet\$1v1\$10.75\$1128 | X | | | X | X | X | | | X | X |
| mobilenet\$1v1\$10.75\$1224 | X | | | X | X | X | | | X | X |
| mobilenet\$1v1\$11.0\$1128 | X | | | X | X | X | | | X | X |
| mobilenet\$1v1\$11.0\$1192 | X | | | X | X | X | | | X | X |
| mobilenet\$1v2\$11.0\$1224 | X | | | X | X | X | | | X | X |
| resnet\$1v2\$1101 | | | | X | X | X | | | X | |
| squeezenet\$12018\$104\$127 | X | | | X | X | X | | | X | |
------
#### [ TensorFlow-Lite (INT8) ]
| Modelos | ARM V8 | ARM Mali | Ambarella CV22 | Nvidia | Panorama | A TDA4VM | Qualcomm QCS603 | X86\$1Linux | X86\$1Windows | i.MX 8M Plus |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| inception\$1v1 | | | | | | | X | | | X |
| inception\$1v2 | | | | | | | X | | | X |
| inception\$1v3 | X | | | | | X | X | | X | X |
| inception\$1v4\$1299 | X | | | | | X | X | | X | X |
| mobilenet\$1v1\$10.25\$1128 | X | | | | | X | | | X | X |
| mobilenet\$1v1\$10.25\$1224 | X | | | | | X | | | X | X |
| mobilenet\$1v1\$10.5\$1128 | X | | | | | X | | | X | X |
| mobilenet\$1v1\$10.5\$1224 | X | | | | | X | | | X | X |
| mobilenet\$1v1\$10.75\$1128 | X | | | | | X | | | X | X |
| mobilenet\$1v1\$10.75\$1224 | X | | | | | X | X | | X | X |
| mobilenet\$1v1\$11.0\$1128 | X | | | | | X | | | X | X |
| mobilenet\$1v1\$11.0\$1224 | X | | | | | X | X | | X | X |
| mobilenet\$1v2\$11.0\$1224 | X | | | | | X | X | | X | X |
| deeplab-v3\$1513 | | | | | | | X | | | |
------