{"id":17195627,"url":"https://github.com/anonymone/ace-nas","last_synced_at":"2025-07-26T05:09:58.293Z","repository":{"id":185514680,"uuid":"179191048","full_name":"anonymone/ACE-NAS","owner":"anonymone","description":null,"archived":false,"fork":false,"pushed_at":"2024-07-25T11:13:47.000Z","size":3413,"stargazers_count":5,"open_issues_count":2,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2025-03-27T11:02:30.706Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/anonymone.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2019-04-03T02:01:36.000Z","updated_at":"2021-07-15T00:42:41.000Z","dependencies_parsed_at":"2023-09-11T04:30:58.211Z","dependency_job_id":null,"html_url":"https://github.com/anonymone/ACE-NAS","commit_stats":null,"previous_names":["anonymone/ace-nas"],"tags_count":7,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anonymone%2FACE-NAS","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anonymone%2FACE-NAS/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anonymone%2FACE-NAS/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anonymone%2FACE-NAS/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/anonymone","download_url":"https://codeload.github.com/anonymone/ACE-NAS/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248781955,"owners_count":21160712,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-10-15T01:51:00.165Z","updated_at":"2025-04-13T20:40:53.443Z","avatar_url":"https://github.com/anonymone.png","language":"Python","readme":"# ACEncoding\nCode accompanying the paper \n\u003e*Action Command Encoding for Surrogate Assisted Neural Architecture Search*\n\nACEncoding is an encoding strategy of NAS. The proposed ACEncoding is inspired by the process of gene expression, where each architecture is constructed by a series of action commands and each action command is encoded by three integers.  \n![ACE](./Res/Figure/ACEncoding.jpg)\n\n## Requirements\n```\npip install -r requirements.txt\n\n# Surrogate dependence\n\ngit clone https://github.com/anonymone/pytorch-seq2seq.git ./Res/seq2seq\n\ncd ./Res/Seq2Seq/\n\npython3 setup.py install\n\n```\n\n## Neural Architecture Search\n```\n# Evolutionary Algorithm\nsh search_EA.sh\n\n# Reinforcement Learning\nsh search_RL.sh\n\n# Gradient Descent\nsh search_GD.sh\n\n# Random (Baseline)\nsh search_RD.sh\n```\n## Neural Architecture Validation\nYour need to specify the paramenter `$ENCODE = \"Your ACEncoding String.\"`\n```\n# CIFAR-10, CIFAR-100, Fashion-MNIST\n# You need to specify the parameter $DATASET = [\"CIFAR10\", \"CIFAR100\", \"Fashion-MNIST\"]\nsh valid_CIFAR.sh\n\n# ImageNet\nsh valid_ImageNet.sh\n```\n\nIf your GPU server has more than one GPU. You can choose the GPU by changing `$DEVICE = cuda:$ID`. `$ID` is the ID of GPUs.\n\n## Visualization\nIn our framework, each neural architecture is saved as a visual DOT file.\n### Install Graphviz\n```\n# CentOS\nsudo yum install graphviz\n\n# Ubuntu\nsudo apt install graphviz\n\n# MacOS\nbrew install graphviz\n```\n### Usage\n```\n# Command Line\ndot -Tpng YOUR_DOT_FILE.dot -o OUTPUT.png\n\n# We highly recommand to use VScode extension Graphviz Preview.\n# And then, you can check your DOT file in a convenience way.\n```\n\n## Acknowledgement\nCode inspired and modified from [NSGA-Net](https://github.com/ianwhale/nsga-net), [NAO](https://github.com/renqianluo/NAO_pytorch), [MetaQNN](https://github.com/bowenbaker/metaqnn), [ENAS](https://github.com/melodyguan/enas) and [IBM/Seq2Seq-torch](https://github.com/IBM/pytorch-seq2seq).","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanonymone%2Face-nas","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fanonymone%2Face-nas","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanonymone%2Face-nas/lists"}