Skip to content

Latest commit

 

History

History
30 lines (27 loc) · 860 Bytes

README.md

File metadata and controls

30 lines (27 loc) · 860 Bytes
Raw

SSD MobileNet v2

MobileNetV2 architecture from "MobileNetV2: Inverted Residuals and Linear Bottlenecks" https://arxiv.org/abs/1801.04381.

For the Pytorch implementation, you can refer to pytorch-ssd

How to run

  1. Generate weights from using gen_wts.py script
git clone https://github.com/qfgaohao/pytorch-ssd.git
cd pytorch-ssd
cp {tensorrtx}/ssd/gen_wts.py .
python gen_wts.py
  1. Put ssdmobilenet.wts into tensorrtx/ssd
cp {pytorch-ssd}/models/ssdmobilenet.wts {tensorrtx}/ssd/ssd_mobilenet.wts
  1. Build and run
cd {tensorrtx}/ssd
mkdir build
cd build
cmake ..
make
sudo ./ssd-mobilenetv2 -s   // serialize model to plan file i.e. 'ssd_mobilenet.engine'
sudo ./ssd-mobilenetv2 -d   // deserialize plan file and run inference
  1. Check if the output is same as {pytorch-ssd}