Skip to content

Latest commit

 

History

History
26 lines (16 loc) · 840 Bytes

README.md

File metadata and controls

26 lines (16 loc) · 840 Bytes

SPAIR_pytorch

Pytorch implementation of Spatially Invariant Unsupervised Object Detection with VAE

Additional information can be found at supplementary paper

Implementation based on the original tensorflow implementation by Eric Crawford: https://github.com/e2crawfo/auto_yolo (Special thanks to Eric for patiently explaining the details of his implementation)

dependencies

  • Numpy
  • Pytorch 1.0+
  • Python 3.5+
  • TensorboardX

Hyperparameters

spair/config.py contains hyperparameters used in the model with comments.

Run

python train.py --gpu to start training with all available GPUs

Data

Coming soon...