RT-DETRv2 train fails to start due to empty Dataloaders

[OfirDataloopAI]

Describe the bug
When I try to train "RT-DETRv2" with "rtdetrv2_pytorch/tools/train.py", I get the following error:

Not init distributed mode.
Start training
Load PResNet101 state_dict
Initial lr: [1e-06, 0.0001, 0.0001]
building train_dataloader with batch_size=16...
loading annotations into memory...
Done (t=0.00s)
creating index...
index created!
building val_dataloader with batch_size=32...
loading annotations into memory...
Done (t=0.00s)
creating index...
index created!
Resume checkpoint from C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_r101vd_6x_coco_from_paddle.pth
Not load model.state_dict
Not load criterion.state_dict
Not load postprocessor.state_dict
Load ema.state_dict
Not load optimizer.state_dict
Not load lr_scheduler.state_dict
Not load lr_warmup_scheduler.state_dict
number of trainable parameters: 76527756
Traceback (most recent call last):
  File "C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_pytorch\tools\train.py", line 69, in <module>
    main(args)
  File "C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_pytorch\tools\train.py", line 35, in main
    solver.fit()
  File "C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_pytorch\src\solver\det_solver.py", line 38, in fit
    train_stats = train_one_epoch(
  File "C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_pytorch\src\solver\det_engine.py", line 38, in train_one_epoch
    for i, (samples, targets) in enumerate(metric_logger.log_every(data_loader, print_freq, header)):
  File "C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_pytorch\src\misc\logger.py", line 238, in log_every
    header, total_time_str, total_time / len(iterable)))
ZeroDivisionError: float division by zero

And I see that the loaders are empty even though the dataset is loaded successfully:

What might be the issue?
It happens for both my train_loader and val_loader.

To Reproduce
I try to train "RT-DETRv2" with "rtdetrv2_pytorch/tools/train.py", and the following json files with 1 item for train and 1 item for validation:

instances_train.json
instances_val.json

And update in the "C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_pytorch\configs\dataset\coco_detection.yml" the fields: "ann_file" and "img_folder" as required.

Then I run the "rtdetrv2_pytorch/tools/train.py" with the following args:

args.config = r"C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_pytorch\configs\rtdetrv2\rtdetrv2_r101vd_6x_coco.yml" args.resume = r"C:\Users\User\PycharmProjects\RT-DETR\rtdetrv2_r101vd_6x_coco_from_paddle.pth"

python env:

absl-py==2.1.0
contourpy==1.3.1
cycler==0.12.1
faster-coco-eval==1.6.5
filelock==3.17.0
fonttools==4.56.0
fsspec==2025.2.0
grpcio==1.70.0
Jinja2==3.1.5
kiwisolver==1.4.8
Markdown==3.7
MarkupSafe==3.0.2
matplotlib==3.10.1
mpmath==1.3.0
narwhals==1.29.0
networkx==3.4.2
numpy==2.2.3
packaging==24.2
pandas==2.2.3
pillow==11.1.0
plotly==6.0.0
protobuf==6.30.0
pycocotools==2.0.8
pyparsing==3.2.1
python-dateutil==2.9.0.post0
pytz==2025.1
PyYAML==6.0.2
scipy==1.15.2
six==1.17.0
sympy==1.13.1
tensorboard==2.19.0
tensorboard-data-server==0.7.2
torch==2.6.0
torchvision==0.21.0
typing_extensions==4.12.2
tzdata==2025.1
Werkzeug==3.1.3