XiaoXYe

Hi @liuxymvp, what's your torch version? MMdnn 0.3.0 only support torch 0.4.0 currently. You can find our requirements in **requirements\linux-py3.txt**

Hi, @johnjwatson ,I have tested your case and it did have an error same as #231 in func **rename_FullyConnected**. But your error locates in func **rename_BatchNorm** and I didn't get...

@johnjwatson ,yes I only changed line 408 in function `rename_FullyConnected` in **mxnet_parser.py** to this : `weight = self.weight_data.get("fc1_weight").asnumpy().transpose((1, 0))` and my package version: ``` mmdnn==0.3.1 mxnet==1.6.0 ```

Hi @ysh329 , it seems there are only 8 groups in your source caffe model and MMdnn convert faithfully

Hi @Manideep08 , slice_axis op is not supported by MXNET parser currently. We will add it in future release.

Hi @budaLi , thanks for your feedback. It seems that pytorch cannot load your model correctly, check if something was wrong with your pth model

Hi @budaLi , this is the trace script of your pytorch model. What about the details of the error's location (python file and function) ? And make sure that your...

Hi @nhatuan84 , thanks for your feedback, we will look into this issue. And do you have the required versions and packages given in the requirements? [https://github.com/microsoft/MMdnn/blob/master/requirements/linux-py3.txt](url)

Hi @Hasan6979 , models downloaded directly from torchvision only have weights (you can check it easily with netron). An extra precedure of load and save is required, which we have...

Hi @Hasan6979 , This code will help you ``` import torch import torchvision.models as models net = models.resnet101(pretrained=False) net.load_state_dict(torch.load('resnet101.pth')) torch.save(net, 'imagenet_resnet101.pth') ```