search

qfgaohao / pytorch-ssd

MobileNetV1, MobileNetV2, VGG based SSD/SSD-lite implementation in Pytorch 1.0 / Pytorch 0.4. Out-of-box support for retraining on Open Images dataset. ONNX and Caffe2 support. Experiment Ideas like CoordConv.
https://medium.com/@smallfishbigsea/understand-ssd-and-implement-your-own-caa3232cd6ad
MIT License
1.39k stars 529 forks source link

ONNX/Caffe2 model conversion issue #136

Open Ashutosh1995 opened 4 years ago

Ashutosh1995 commented 4 years ago

Hi,

Thanks for the repo containing all the models!

While converting pretrained mobilnet v1 SSD model to caffe2 and onnx, there is no predictor layer attached at the end because of which we only supply outputs as scores and boxes and miss out labels in the onnx conversion file.

How should I get a complete model in onnx which also has a predictor head and gives class names at the output ?

shashwat623 commented 3 years ago

Hi,

Thanks for the repo containing all the models!

While converting pretrained mobilnet v1 SSD model to caffe2 and onnx, there is no predictor layer attached at the end because of which we only supply outputs as scores and boxes and miss out labels in the onnx conversion file.

How should I get a complete model in onnx which also has a predictor head and gives class names at the output ?

@Ashutosh1995 I'm facing the same issue. Any updates ?