Skip to content

Latest commit

 

History

History

gates

README.md

A Generic Graph-based Neural Architecture Encoding Scheme for Predictor-based NAS (GATES), in ECCV 2020

If you find this work/repo helpful, please cite:

@misc{ning2020generic,
    title={A Generic Graph-based Neural Architecture Encoding Scheme for Predictor-based NAS},
    author={Xuefei Ning and Yin Zheng and Tianchen Zhao and Yu Wang and Huazhong Yang},
    year={2020},
    eprint={2004.01899},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

The implementations of GATES (detailed in Appendix. Sec.1) in OON/OOE search spaces are in aw_nas/utils/torch_utils.py: DenseGraphFlow: OON, e.g., NAS-Bench-101; DenseGraphOpEdgeFlow: OOE, e.g., ENAS; DenseGraphSimpleOpEdgeFlow: simple OOE, e.g., NAS-Bench-201.

We provide the sample codes/logs/configs of training different predictors on NAS-Bench-101, and the sample logs/configs of trainig different predictors on NAS-Bench-201 (Corresponding to the Sec. 4.1 ~ Sec. 4.2 in the paper). The sample configs/logs are in two directories: nb101_cfgs, and nb201_cfgs from this url.

  • Under nb101_cfgs/tr_5e-4 are the example configurations and logs on NAS-Bench-101. 0.05% (190) of the training data are used to train the predictors. You could run bash print_res.sh nb101_cfgs to print the average test kendall's tau of the last 5 epochs of these exps.
  • Under nb201_cfgs/ are the example configurations and logs on NAS-Bench-201. 1%/10% of the training data are used to train the predictors. You could run bash print_res.sh nb201_cfgs to print the average test kendall's tau of the last 5 epochs of these exps.

You can also use the sample aw_nas code provided in the cloud drive to run the test, while this repository provides the full aw_nas codes.

Run the training exps on NAS-Bench-101

Folow the following steps to run experiments.

Install

Dump nb101 data

Dump nb101 pickle data (all patched to 7x7):

python dump_nb101_allv.py

Two pickle files would be generated under the current working dir: nasbench_allv.pkl, nasbench_allv_valid.pkl

Modify configurations

Optionally, make some changes to the config.yaml files under nb101_cfgs/tr_5e-4, for example

  • To use 1% of the training archs rather than 0.05% (190). Modify train_ratio: 0.0005 to train_ratio: 0.01 in these configs.
  • To use the regression loss instead of the hinge ranking loss, modify compare: true to compare: false in these configs.
  • To use different number of GATES/GCN layers, change the arch_network_cfg.arch_embedder_cfg.gcn_out_dims list.
  • See the config file for more configurations!

Run exp

Run python scripts/nasbench/train_nasbench_pkl.py <yaml_config_file> --train-dir <train_result_dir> --gpu 0