ShuffleNet V2

Use case : Image classification

Model description

ShuffleNet V2 is designed following practical guidelines for efficient CNN architecture design. It uses channel shuffle operations and a split-concat structure for efficient feature reuse with minimal memory access cost.

The architecture features channel shuffle operations to enable information flow between channel groups, with a split-concat architecture for efficient feature processing. Designed based on practical guidelines using direct speed measurement rather than FLOPs, the architecture makes choices that minimize memory access cost.

ShuffleNet V2 is well-suited for mobile applications with strict efficiency requirements, real-time video processing, and multi-model deployment scenarios.

(source: https://arxiv.org/abs/1807.11164)

The model is quantized to int8 using ONNX Runtime and exported for efficient deployment.

Network information

Network inputs / outputs

For an image resolution of NxM and P classes

Recommended platforms

Performances

Metrics

• Measures are done with default STEdgeAI Core configuration with enabled input / output allocated option.
• All the models are trained from scratch on Imagenet dataset

Reference NPU memory footprint on Imagenet dataset (see Accuracy for details on dataset)

Reference NPU inference time on Imagenet dataset (see Accuracy for details on dataset)

Accuracy with Imagenet dataset

Retraining and Integration in a simple example:

Please refer to the stm32ai-modelzoo-services GitHub here

References

[1] - Dataset: Imagenet (ILSVRC 2012) — https://www.image-net.org/

[2] - Model: ShuffleNet V2 — https://github.com/megvii-model/ShuffleNet-Series