HardNet

Use case : Image classification

Model description

Harmonic DenseNet (HardNet) is a memory-efficient variant of DenseNet that optimizes for both computational efficiency and memory access cost. It introduces a harmonic pattern in the dense connections to reduce redundant feature computations.

HardNet features harmonic dense connections that reduce connection patterns to minimize memory bandwidth, while maintaining the benefits of DenseNet's feature reuse. The architecture combines depthwise separable convolutions with dense blocks for enhanced efficiency.

Designed for practical hardware deployment, HardNet provides DenseNet-like feature richness with lower memory cost on edge devices.

(source: https://arxiv.org/abs/1909.00948)

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

Dataset details: link Number of classes: 1000. To perform the quantization, we calibrated the activations with a random subset of the training set. For the sake of simplicity, the accuracy reported here was estimated on the 50000 labelled images of the validation set.

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: HarDNet — https://github.com/PingoLH/Pytorch-HarDNet