OpenVino的MXnet模型转换
[在使用Movidius的模型优化器转换模型之前,需要先用MXNet的deploy.py将模型转换成部署模式,然后才能用movidius的优化器转换]
https://github.com/apache/incubator-mxnet/blob/master/example/ssd/deploy.py
cd ~
git clone https://github.com/apache/incubator-mxnet
mv tmp/*-0000.params tmp/ssd_resnet50_512-0000.params
mv tmp/*-symbol.json tmp/ssd_resnet50_512-symbol.json
python3 incubator-mxnet/example/ssd/deploy.py --network resnet50 --data-shape 512 --num-class 5 --prefix tmp/ssd_
cd /opt/intel/openvino/deployment_tools/model_optimizer
python3 mo_mxnet.py --input_model ssd_resnet50_512-0000.params --input_shape [1,3,512,512]
python3 mo_mxnet.py --input_model ssd_resnet50_512-0000.params --input_shape [1,3,512,512] --data_type=FP16