[Trivial Fix] Flag use_ema in train_unconditional.py should not have default value

[nkirschi]

Problem

The definition of the flag use_ema wrongly includes a default value of True:

diffusers/examples/unconditional_image_generation/train_unconditional.py

Lines 156 to 161 in fc94c60

	parser.add_argument(
	"--use_ema",
	action="store_true",
	default=True,
	help="Whether to use Exponential Moving Average for the final model weights.",
	)

Hence, when the flag is not passed via CLI, it is still evaluated as true.

Fix

To fix this bug, simply remove default=True from the flag definition.

[patrickvonplaten]

Agree actually, we should probably rename this to --use_no_ema?

cc @anton-l

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.