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.
