Exception in train.py - Can't pickle local object 'train_detector.<locals>.<lambda>

[johannesTech]

Hello @Teoge,
first of all, thank you for sharing your research together with your code here!

I tried to execute your code and the inference ran flawlessly. Unfortunately, the training is not running due to the following error at location

DMPR-PS/train.py

Lines 73 to 79 in 88b77e3

	data_loader = DataLoader(data.ParkingSlotDataset(args.dataset_directory),
	batch_size=args.batch_size, shuffle=True,
	num_workers=args.data_loading_workers,
	collate_fn=lambda x: list(zip(*x)))
	for epoch_idx in range(args.num_epochs):
	for iter_idx, (images, marking_points) in enumerate(data_loader):

Exception has occurred: AttributeError Can't pickle local object 'train_detector.<locals>.<lambda>'
This issue is seems to be a result of the lambda function being called in the DataLoader.

I executed your code with python 3.7.0 and torch 1.7.0.

I would appreciate any hint you have for me :)

Best,
Johannes

[Teoge]

Did you run your training on windows? It looks similar to this issue.
You can try the solutions inside, replacing lambda with a normally defined function, or setting num_workers to 0.