[python] support Dataset.get_data for Sequence input. #4472
Conversation
cyfdecyf
requested review from
chivee,
henry0312,
jameslamb,
shiyu1994 and
StrikerRUS
as code owners
There was a problem hiding this comment.
@cyfdecyf Many thanks for this PR! Please check my initial thoughts below:
cyfdecyf
force-pushed
the
sequence-dataset
branch
from
160c1de to
37b781b
Compare
| elif isinstance(self.data, Sequence): | ||
| self.data = self.data[self.used_indices] | ||
| elif isinstance(self.data, list) and len(self.data) > 0 and all(isinstance(x, Sequence) for x in self.data): | ||
| self.data = np.array([row for row in self._yield_row_from_seqlist(self.data, self.used_indices)]) |
There was a problem hiding this comment.
Please add test cases with Sequence and List[Sequence] for add_features_from() here
There was a problem hiding this comment.
Also, please enhance this test (or just add a new one without a parametrization to save some CI time)
with new
get_data() functionality for Sequence(s) raw data.
There was a problem hiding this comment.
Thanks for the review. I'll add test cases later this week.
Co-authored-by: Nikita Titov <nekit94-08@mail.ru>
|
Test cases are added now. It's a little hacky because I'm not familiar with the use cases. |
There was a problem hiding this comment.
@cyfdecyf Thank you very much for continuing to enhance Sequence support!
Test cases are added now. It's a little hacky because I'm not familiar with the use cases.
No problem! I can slightly refactor them later. Don't want to delay this PR.
|
This pull request has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this. |
Implement features for #4403.