Hi @ibaiGorordo ,

Thank you for your great work !.

I've tried to convert MobileStereoNet Pytorch model to ONNX model by using torch onnx export. However the ONNX model size is 10 times bigger than Pytorch model size. (i.e. 300Mbs, compare to 29Mbs).

The script which I used to convert to onnx model:

import torch
import onnx
from onnxsim import simplify
from models import MSNet2D

if __name__ == '__main__':

    # Load models
    maxdisp = 192

    model = MSNet2D(maxdisp)
    model.cuda()

    print('load model')
    model_path = 'weights/MSNet2D_SF_KITTI2015.ckpt'
    state_dict = torch.load(model_path)

    pretrained_dict = {key.replace("module.", ""): value for key, value in state_dict['model'].items()}
    model.load_state_dict(pretrained_dict)
    model.eval()
  
    W = 528
    H = 240
    onnx_file = f"mobileStereoNet_{H}x{W}.onnx"
    onnx_simp_file = f"mobileStereoNet_{H}x{W}_simp.onnx"
    
    x = torch.randn(1, 3, H, W).cuda()
    torch.onnx.export(model,
                                args=(x, x),
                                f=onnx_file,
                                opset_version=12)

    print("save model {} successfully".format(onnx_file))

    model_load = onnx.load(onnx_file)
    model_simp, check = simplify(model_load)
    onnx.save(model_simp, onnx_simp_file)

    print("save model {} successfully".format(onnx_simp_file))

Could you hint me how did you convert MobileStereoNet Pytorch to ONNX ?

Many thanks in advance !.

Regards, Michael

Firstly, thank you for providing excellent work.

I was about to work on converting the mobilestereonet onnx model to TensorRT model to improve inference speed. Even though I confirmed my own model was well converted to onnx using your work, but issues come up as I tried to convert the onnx to a TensorRT model.

Have you tried the conversion? If so, it will be great if you could share your process since I'm struggling with various issues. I've tried TRT6 in Ubuntu18.04, CUDA10.1 environment.

Thanks again!