kunglab / ddnn Goto Github PK

module' object has no attribute '_check_cudnn_acceptable_type

Hi, I have troubles about understanding the code..

1. I can't find the part checking whether the inference accuracy is higher than threshold
2. I want to know how to communicate between end devices and cloud.

thank you :)

Hi, I am interested in your work and find the paper writes DDNN supports the end-edge-cloud computatoin distribution. But in the code I only find the models on the end and the cloud. Could you please help:

1. I find the model on the end and the cloud differs from channel numbers, can I generalize this feature if the hierachy is end-edge-cloud.
2. In end-cloud computation, the cloud gathers multi-view data from end devices. If the edge works as the middle level, is the offloading from edge to cloud one-to-one offloading?

Hi
I have problems running the code due to an error created by the OS: "The paging file is too small for this operation to complete. Error loading "C:\python\Python36\lib\site-packages\torch\lib\caffe2_detectron_ops_gpu.dll" or one of its dependencies."
The environment I'm using is PyCharm using python 3.6 interpreter and torch 1.10.2 + cuda11 (wasn't able to download previous version).
Can you help handle this error ?
Thank you

how can i start it?

Dear author,
I am interested in your research.
I want to know what is the DDNN architecture according to the paper named Distributed Deep Neural Networks over the Cloud, the Edge and End Devices. In this paper,you use two types of blocks: FC and ConvP, so I also want know which part of the code represents FC block and Convp respectively.

Hi : I have problems about how to run this code

1. Which code will run in the end device ?
2. Which code is about the communication between end devices and cloud?

Dear author，
I am also interested in your research.
Nearly a month passed,I want to know will you upload a minimal example for us to reproduce the experiment? If so,what's the timeline?

Thanks,

Dear authors,

I'm interested in reproducing the results in your paper.
Could you provide a quick tutorial on how to reproduce the experiments in the paper?

Thanks,