Home

Awesome

PNASNet.pytorch

PyTorch implementation of PNASNet-5. Specifically, PyTorch code from this repository is adapted to completely match both my implemetation and the official implementation of PNASNet-5, both written in TensorFlow. This complete match allows the pretrained TF model to be exactly converted to PyTorch: see convert.py.

If you use the code, please cite:

@inproceedings{liu2018progressive,
  author    = {Chenxi Liu and
               Barret Zoph and
               Maxim Neumann and
               Jonathon Shlens and
               Wei Hua and
               Li{-}Jia Li and
               Li Fei{-}Fei and
               Alan L. Yuille and
               Jonathan Huang and
               Kevin Murphy},
  title     = {Progressive Neural Architecture Search},
  booktitle = {European Conference on Computer Vision},
  year      = {2018}
}

Requirements

Data and Model Preparation

python convert.py

Notes on Model Conversion

Image SizeOfficial TensorFlow ModelConverted PyTorch Model
(331, 331)(0.829, 0.962)(0.828, 0.961)
(323, 323)(0.827, 0.961)(0.827, 0.961)

Usage

python main.py

The last printed line should read:

Test: [50000/50000]	Prec@1 0.828	Prec@5 0.961