mnasnet

MNASNet with depth multiplier of 0.5 from "MnasNet: Platform-Aware Neural Architecture Search for Mobile" https://arxiv.org/pdf/1807.11626.pdf

For the Pytorch implementation, you can refer to pytorchx/mnasnet

Following tricks are used in this mnasnet, nothing special, group conv and batchnorm are used.

Batchnorm layer, implemented by scale layer.

Usage

use gen_wts.py to generate wts file

bash

python gen_wts.py

build C++ code

bash

pushd tensorrtx/mnasnet
cmake -S . -B build -G Ninja --fresh
cmake --build build

serialize wts model to engine file

bash

./build/mnasnet -s

run inference

bash

./build/mnasnet -d

The output looks like:

bash

...
====
Execution time: 0ms
-2.024, -1.266, -1.602, -1.465, -0.7756, -0.2096, 0.05945, 1.342, -0.2382, 1.279, 1.251, 0.2579, 1.836, -0.5296, 0.3196, 0.9055, -0.4915, 0.1604, -0.6305, -0.1019, -0.8816,
====
prediction result:
Top: 0 idx: 285, logits: 4.869, label: Egyptian cat
Top: 1 idx: 281, logits: 4.837, label: tabby, tabby cat
Top: 2 idx: 282, logits: 4.019, label: tiger cat