Signed-off-by: Wenqi Li [email protected]

Description

follow-up of #1833

this is a quick implementation to provide more info when collating fails.
if it's useful I can finish this PR with more tests and options

import numpy as np
import torch
from monai.data.utils import dev_collate

dev_collate(
    [
        {"img": 2, "meta": {"shape": [torch.tensor(1.0)]}},
        {"img": 3, "meta": {"shape": [np.asarray(1.0)]}},
        {"img": 4, "meta": {"shape": [torch.tensor(1.0)]}},
    ]
)

logging diagnostic info:

> collate dict key "img" out of 2 keys
> collate dict key "meta" out of 2 keys
>> collate dict key "shape" out of 1 keys
>>> collate list of sizes: [1, 1, 1].
>>>> collate/stack a list of tensors
>>>> E: expected Tensor as element 1 in argument 0, but got numpy.ndarray, while stacking ['Tensor', 'ndarray', 'Tensor'] in collate((tensor(1.), array(1.), tensor(1.)))

Status

Ready

Types of changes

• Non-breaking change (fix or new feature that would not break existing functionality).
• Breaking change (fix or new feature that would cause existing functionality to change).
• New tests added to cover the changes.
• Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
• Quick tests passed locally by running ./runtests.sh --quick --unittests --disttests.
• In-line docstrings updated.
• Documentation updated, tested make html command in the docs/ folder.