Csharp data preprocessing about code2seq HOT 5 CLOSED

Closing due to inactivity, feel free to re-open if you have further questions!

from code2seq.

Hi @greenmonn ,
I tested this with dotnet core 2.2 .
But let's try running the dotnet command directly. Can you please run:

dotnet run --project CSharpExtractor/CSharpExtractor/Extractor/Extractor.csproj --max_length 8 --max_width 2 --path CSharpExtractor/CSharpExtractor --threads 64 --ofile_name my_dataset.val.raw.txt

And tell me what it outputs?

from code2seq.

Thanks for the reply!
Now I realized it was just the problem of the specified dotnet version. When I ran the command, I got the below result:

After reinstalling dotnet-sdk with version 2.2, now datasets are successfully preprocessed. However, I think it might work okay with 3.1 version by removing the specified version part on CExtractor.. Anyway, thanks for the great repo :)

from code2seq.

By the way, I have a one more question. I tried to use stackoverflow dataset of CodeNN, but I only found some .txt files of parsed code snippets that were already preprocessed for their model, from the CodeNN repository. Could you give some advice to correctly replicate the result using the stackoverflow dataset?

from code2seq.

We did not release official adaptation preprocessing for the CodeNN dataset, because we found it to be very noisy and I don't recommend using this dataset.
See also the discussion and pointers here: #17

from code2seq.