guberti commented on code in PR #12521:
URL: https://github.com/apache/tvm/pull/12521#discussion_r952400710
##########
python/tvm/autotvm/task/dispatcher.py:
##########
@@ -30,18 +30,26 @@
from __future__ import absolute_import as _abs
+from io import TextIOBase
import logging
-import typing
-from typing import Union
-from collections.abc import Iterable
+from os import PathLike
+from pathlib import Path
+from typing import List, Iterable, Tuple, Union
import numpy as np
from .space import FallbackConfigEntity
from .. import env as _env
+from ..measure import MeasureInput, MeasureResult
logger = logging.getLogger("autotvm")
+Records = Union[
+ Union[str, bytes, Path], # Path-like objects
+ TextIOBase, # File-like objects
+ Iterable[Tuple[MeasureInput, MeasureResult]],
Review Comment:
Yes, `ApplyHistoryBest` currently supports taking an input of type
`Iterable[Tuple[MeasureInput, MeasureResult]]`. This functionality is basically
never used though, as autotuning (AFAIK) does not support exporting logs with
this type. Do you think this functionality can be removed?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
