Datasets:
Tasks:
Text Generation
Sub-tasks:
language-modeling
Languages:
code
Size:
10K<n<100K
ArXiv:
Tags:
code
License:
Empty "solutions" in train dataset
#3
by
lychees
- opened
It seems that there is more than >5000 items without "solutions" data in train dataset.
And "url" is missing for atcoder, hackerrank && aizu.
It seems that there is more than >5000 items without "solutions" data in train dataset.
TACO can provides (code problem, solution) pairs for code generation and (code problem, algorithm tag) pairs for code analysis. So there are some problems without solutions which are remained.
And "url" is missing for atcoder, hackerrank && aizu.
The atcoder data are from APPS/CodeContests, and the "url" content is also from these two datasets.
Thanks for your attention and advice for our work and if you have other questions, please feel free to contact us.
bowen92
changed discussion status to
closed