search

SyGoing / LFFD-OpenVINO

Intel OpenVINO Implementation of the paper of LFFD: A Light and Fast Face Detector for Edge Devices
45 stars 7 forks source link

Conversion Error #3

Open amromidhat opened 4 years ago

amromidhat commented 4 years ago

Dear SyGoing, I have been trying to convert lffd models to openvino, I have edited the symbol .json as stated but I keep facing this error :

[ ERROR ] Input model name /home/avidbeam/amrm_workspace/lffd_openvino/v4_openvino/train_30_320_20L_4scales_v1_iter_500000.params is not in an expected format, cannot extract iteration number.

here is my conversion command: python3 mo_mxnet.py --input_symbol ./v4_openvino_deploy.json --input_model ./v4_iter_320000.params

Do you have any hints for a successful conversion? Thanks in advance

SyGoing commented 4 years ago

Hi , Maybe you could try to convert the mxnet model to onnx form firstly, then convert onnx to openvino. I have checked it and it works well.