ternaus/check_orientation

Model to check if image was rotated by 90, 180, 270 degrees.

Check orientation

Models to check if image was rotated by 0, 90, 180, 270 degrees.

Installation

pip install -U check_orientation

Example inference

Colab notebook with the example:

Training

Define the config.

Example at check_orientation/configs

Define the environmental variable TRAIN_IMAGE_PATH that points to the folder with train dataset.

Example:

export TRAIN_IMAGE_PATH=<path to the tranining folder>

Define the environmental variable VAL_IMAGE_PATH that points to the folder with validation dataset.

Example:

export VAL_IMAGE_PATH=<path to the validation folder>

Training

python -m check_orientation.train -c <path to config>

Inference

python -m torch.distributed.launch --nproc_per_node=<num_gpu> check_orientation/inference.py \
                                   -i <path to images> \
                                   -c <path to config> \
                                   -w <path to weights> \
                                   -o <output-path> \
                                   --fp16

Pre-trained models

Models were pre-trained on the OpenImages dataset.

Models	Validation accuracy	Config file	Weights
swsl_resnext50_32x4d	0.9128	Link	Link