retinaface(mxnet) to tensorflow error

[Zheweiqiu]

Platform (like ubuntu 16.04/win10): 16.04

Python version: 3.7

Source framework with version (like Tensorflow 1.4.1 with GPU): mxnet 1.5 with GPU

Destination framework with version (like CNTK 2.3 with GPU): tensorflow 1.14 with GPU

Pre-trained model path (webpath or webdisk path): R50-symbol.json R50-0000.params

Running scripts: python -m mmdnn.conversion._script.convertToIR -f mxnet -n R50-symbol.json -w R50-0000.params -d retineface --inputShape 3,640,640

After running above command, I successfully get the .json, .pb, .npy file, but the size of .npy seems not right.

size of R50-0000.params: 113M size of .npy: 496K

Any help will be appreciate!

[rainLiuplus]

Hi @Zheweiqiu, is this the model you use?

[Zheweiqiu]

Hi @Zheweiqiu, is this the model you use?

I am using pre-trained retinaface model from [https://github.com/deepinsight/insightface/tree/master/RetinaFace]

[cydawn]

I meet the same problem with you. Did you fix it? @Zheweiqiu @rainLiuplus

[xandox]

Hi @cydawn , @Zheweiqiu , @rainLiuplus Do you solve this issue?