yihanjiang / turboae Goto Github PK

I check the source code and the paper, and I find that the channels are all assumed as real-value. So I wonder if I try to use the complex channels which is practical actually, does it work?

for example, the following error appears when I want to test pre-trained model as you said in readme file.

`CUDA_VISIBLE_DEVICES=0 : The term 'CUDA_VISIBLE_DEVICES=0' is not recognized as the name of a cmdlet, function, script
file, or operable program. Check the spelling of the name, or if a path was included, verify that the path is correct
and try again.
At line:1 char:1

• CUDA_VISIBLE_DEVICES=0 python3.6 main.py -encoder TurboAE_rate3_cnn - ...

•   + CategoryInfo          : ObjectNotFound: (CUDA_VISIBLE_DEVICES=0:String) [], CommandNotFoundException
    + FullyQualifiedErrorId : CommandNotFoundException`
  

Hi Yihan, could you please tell me if your implementation for the ldpc/polar/turbo shown in the Figure 1 are using the code from the "commpy" folder on your github?

Thanks in advance!

I reach to the following error.

PS C:\Users\Asus\Desktop\idea\FTN\Deep turbo\turboae-master> & C:/Users/Asus/AppData/Local/Programs/Python/Python311/python.exe "c:/Users/Asus/Desktop/idea/FTN/Deep turbo/turboae-master/commpy/examples/conv_encode_decode.py" File "c:\Users\Asus\Desktop\idea\FTN\Deep turbo\turboae-master\commpy\examples\conv_encode_decode.py", line 14 generator_matrix = np.array([[05, 07]]) ^ SyntaxError: leading zeros in decimal integer literals are not permitted; use an 0o prefix for octal integers PS C:\Users\Asus\Desktop\idea\FTN\Deep turbo\turboae-master>