Visitar URL original
Omit `dataloader_kwargs` from `TrainSettingsConfig` if unset to ensure forward compatibility by deruyter92 · Pull Request #3576 · DeepLabCut/DeepLabCut · GitHub
Skip to content

Omit dataloader_kwargs from TrainSettingsConfig if unset to ensure forward compatibility #3576

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
C-Achard merged 6 commits into from
Nov 5, 2026
Merged

Omit dataloader_kwargs from TrainSettingsConfig if unset to ensure forward compatibility #3576

Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
eb32970
add model_serializer for TrainSettingsConfig.dataloader_kwargs
deruyter92 Nov 3, 2026
cf607d8
dataset_kwargs: Use pydantic 2.11 `exclude_if` instead of model seria…
deruyter92 Nov 3, 2026
abdb724
update pyproject.toml + uv lock: require pydantic 2.11
deruyter92 Nov 3, 2026
d8838ce
Update pyproject.toml
deruyter92 Nov 3, 2026
f63c524
Merge remote-tracking branch 'upstream/dev' into jaap/forward-compati…
deruyter92 Nov 4, 2026
55344ad
update uv.lock
deruyter92 Nov 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
dataset_kwargs: Use pydantic 2.11 exclude_if instead of model seria… 
…lizer
  • Loading branch information
@deruyter92
deruyter92 committed Nov 3, 2026
commit cf607d86fcbc447650cc9e7951c906ca69c2fecf
13 changes: 3 additions & 10 deletions deeplabcut/pose_estimation_pytorch/config/training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@

from typing import Any

from pydantic import SerializerFunctionWrapHandler, field_validator, model_serializer
from pydantic import Field, field_validator

from deeplabcut.core.config import DLCBaseConfig
from deeplabcut.core.config.validation import reject_reserved_keys
Expand Down Expand Up @@ -49,7 +49,8 @@ class TrainSettingsConfig(DLCBaseConfig):
batch_size: int = 8
dataloader_workers: int = 0
dataloader_pin_memory: bool = False
dataloader_kwargs: dict[str, Any] | None = None
# Omitted when empty to keep saved configs loadable by older DeepLabCut versions
dataloader_kwargs: dict[str, Any] | None = Field(default=None, exclude_if=lambda v: not v)
display_iters: int = 500
epochs: int = 200
seed: int = 42
Expand All @@ -62,11 +63,3 @@ class TrainSettingsConfig(DLCBaseConfig):
def validate_dataloader_kwargs(cls, v: dict[str, Any] | None) -> dict[str, Any] | None:
reject_reserved_keys(v, _RESERVED_DATALOADER_KWARGS, name="dataloader_kwargs")
return v

@model_serializer(mode="wrap")
def _omit_empty_dataloader_kwargs(self, handler: SerializerFunctionWrapHandler) -> dict[str, Any]:
# Keep configs without dataloader_kwargs loadable by older DeepLabCut versions
data = handler(self)
if not data.get("dataloader_kwargs"):
data.pop("dataloader_kwargs", None)
return data