Is your feature request related to a problem? Please describe.
Currently, we usually use output_transform=lambda x: None to avoid logging of every iteration for evaluator:
https://github.com/Project-MONAI/tutorials/blob/master/modules/engines/unet_training_dict.py#L126
It's hard to understand, already got several questions from NVIDIA Clara MMAR users.

Would be nice to enhance from_engine API or add another similar API to improve the user experience:
https://github.com/Project-MONAI/tutorials/blob/master/modules/engines/unet_training_dict.py#L162