DETR model for multi-mnist dataset The directory structure of the code is as follws . ├── datasets │ ├── __init__.py │ ├── mmnist │ │ ├── test │ │ │ └── normal │ │ │ ├── bboxes.pickle │ │ │ ├── filenames.pickle │ │ │ ├── imgs │ │ │ └── labels.pickle │ │ └── train │ │ └── normal │ │ ├── bboxes.pickle │ │ ├── filenames.pickle │ │ ├── imgs │ │ └── labels.pickle │ └── mmnist.py ├── DETR_mmnist_inference.ipynb ├── DTER_mmnist_train.ipynb ├── models │ ├── backbone.py │ ├── detr.py │ ├── __init__.py │ ├── __init__.pyc │ ├── matcher.py │ ├── position_encoding.py │ ├── transformer_attn.py │ ├── transformer.py │ └── util │ ├── box_ops.py │ ├── misc.py │ └── __pycache__ │ ├── box_ops.cpython-36.pyc │ └── misc.cpython-36.pyc ├── README └── utils.py Need pytorch 1.5 or later python2. Please refer requirements.txt for further detail. The inference code is written in DETR_mmnist_inference.ipynb The training code is written in DTER_mmnist_train.ipynb The model contains the same code as given in official facebook research repository. The datasets contain code for the dataloader.