Skip to content

Latest commit

 

History

History
112 lines (80 loc) · 3.88 KB

README.md

File metadata and controls

112 lines (80 loc) · 3.88 KB

AtomNAS: Fine-Grained End-to-End Neural Architecture Search [PDF]

Updates

  • [Mar 2020] A clean mobilenet-series implementation is provided.
  • [Feb 2020] Simplify validation process, released the pretrained models. Conflict with previous version.

Overview

This is the codebase (including search) for ICLR 2020 paper AtomNAS: Fine-Grained End-to-End Neural Architecture Search.

Setup

Distributed Training

Set the following ENV variable:

$DATA_ROOT: Path to data root
$METIS_WORKER_0_HOST: IP address of worker 0
$METIS_WORKER_0_PORT: Port used for initializing distributed environment
$METIS_TASK_INDEX: Index of task
$ARNOLD_WORKER_NUM: Number of workers
$ARNOLD_WORKER_GPU: Number of GPUs (NOTE: should exactly match local GPU numbers with `CUDA_VISIBLE_DEVICES `)
$ARNOLD_OUTPUT: Output directory

Non-Distributed Training (Not Recommend)

Set the following ENV variable:

$DATA_ROOT: Path to data root
$ARNOLD_WORKER_GPU: Number of GPUs (NOTE: should exactly match local GPU numbers with `CUDA_VISIBLE_DEVICES `)
$ARNOLD_OUTPUT: Output directory

Reproduce AtomNAS results

For Table 1

  • AtomNAS-A: bash scripts/run.sh apps/slimming/shrink/atomnas_a.yml
  • AtomNAS-B: bash scripts/run.sh apps/slimming/shrink/atomnas_b.yml
  • AtomNAS-C: bash scripts/run.sh apps/slimming/shrink/atomnas_c.yml

If everything is OK, you should get similar results.

Pretrained Models could be downloaded from onedrive

Testing

For AtomNAS:

FILE=$(realpath {{log_dir_path}}) checkpoint=ckpt ATOMNAS_VAL=True bash scripts/run.sh apps/eval/eval_shrink.yml

For AtomNAS+:

TRAIN_CONFIG=$(realpath {{train_config_path}}) ATOMNAS_VAL=True bash scripts/run.sh apps/eval/eval_se.yml --pretrained {{ckpt_path}}

Related Info

  1. Requirements

    • See requirements.txt
  2. Environment

    • The code is developed using python 3. NVIDIA GPUs are needed. The code is developed and tested using 4 servers with 32 NVIDIA V100 GPU cards. Other platforms or GPU cards are not fully tested.
  3. Dataset

    • Prepare ImageNet data following pytorch example.
    • Optional: Generate lmdb dataset by utils/lmdb_dataset.py. If not, please overwrite dataset:imagenet1k_lmdb in yaml to dataset:imagenet1k.
    • The directory structure of $DATA_ROOT should look like this:
      ${DATA_ROOT}
      ├── imagenet
      └── imagenet_lmdb
      
  4. Miscellaneous

    • The codebase is a general ImageNet training framework using yaml config with several extension under apps dir, based on PyTorch.
      • YAML config with additional features
        • ${ENV} in yaml config.
        • _include for hierachy config.
        • _default key for overwriting.
        • xxx.yyy.zzz for partial overwriting.
      • --{{opt}} {{new_val}} for command line overwriting.

Acknowledgment

This repo is based on slimmable_networks and benefits from the following projects

Thanks the contributors of these repos!

Citation

If you find this work or code is helpful in your research, please cite:

@inproceedings{
    mei2020atomnas,
    title={Atom{NAS}: Fine-Grained End-to-End Neural Architecture Search},
    author={Jieru Mei and Yingwei Li and Xiaochen Lian and Xiaojie Jin and Linjie Yang and Alan Yuille and Jianchao Yang},
    booktitle={International Conference on Learning Representations},
    year={2020},
    url={https://openreview.net/forum?id=BylQSxHFwr}
}