{"id":33011920,"url":"https://github.com/microsoft/petridishnn","last_synced_at":"2025-11-18T05:00:35.964Z","repository":{"id":41284082,"uuid":"180651701","full_name":"microsoft/petridishnn","owner":"microsoft","description":"Code for the neural architecture search methods contained in the paper Efficient Forward Neural Architecture Search","archived":false,"fork":false,"pushed_at":"2023-06-12T18:22:32.000Z","size":493,"stargazers_count":110,"open_issues_count":2,"forks_count":21,"subscribers_count":12,"default_branch":"master","last_synced_at":"2025-10-20T02:42:46.660Z","etag":null,"topics":["automl","cifar-100","cifar10","deep-learning","deep-learning-algorithms","image-classification","imagenet","neural-architecture-search"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/microsoft.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2019-04-10T19:39:36.000Z","updated_at":"2025-09-23T22:41:14.000Z","dependencies_parsed_at":"2024-01-12T17:35:42.623Z","dependency_job_id":"c22dfbf8-b2d0-43ae-ba22-707a95d16a2e","html_url":"https://github.com/microsoft/petridishnn","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/microsoft/petridishnn","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/microsoft%2Fpetridishnn","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/microsoft%2Fpetridishnn/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/microsoft%2Fpetridishnn/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/microsoft%2Fpetridishnn/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/microsoft","download_url":"https://codeload.github.com/microsoft/petridishnn/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/microsoft%2Fpetridishnn/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":284621240,"owners_count":27036418,"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","status":"online","status_checked_at":"2025-11-15T02:00:06.050Z","response_time":57,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["automl","cifar-100","cifar10","deep-learning","deep-learning-algorithms","image-classification","imagenet","neural-architecture-search"],"created_at":"2025-11-13T17:00:24.937Z","updated_at":"2025-11-15T21:01:07.696Z","avatar_url":"https://github.com/microsoft.png","language":"Python","readme":"# Project Petridish: Efficient Forward Architecture Search\n----------------------\nWARNING DEPRECATED! DEPRECATED! DEPRECATED!\nMuch higher quality, performant code for Petridish is now available [here](https://github.com/microsoft/archai/blob/master/docs/petridish.md)\nThis repository is not maintained or supported anymore.\n----------------------\n\nCode for [Efficient Forward Neural Architecture Search](https://arxiv.org/abs/1905.13360), Neurips 2019!\n\n*Note this repo is under active development and the code base is expected to rapidly change.*\n*We are currently rewriting Petridish in Pytorch with evaluation on many more datasets and pretrained models. It will appear here shortly.*\n\n## Conduct and Privacy\nPetridishnn has adopted the Microsoft [Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct). For more information on this code of conduct, see the [Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/) or contact opencode@microsoft.com with any additional questions or comments. Read Microsoft’s statement on [Privacy \u0026 Cookies](https://privacy.microsoft.com/en-us/privacystatement/)\n\n# Installation on development machine\nWe have developed and tested Petridish on Ubuntu 16.04 LTS (64-bit), Anaconda python distribution and Tensorflow.\n\n## Installing the software\n1. Install [Anaconda python distribution for Ubuntu](https://www.anaconda.com/distribution/)\n2. Create a python 3.6 environment `conda create python=3.6 -n py36`\n3. Follow instructions to install a recent [Tensorflow (TF) version](https://www.tensorflow.org/install). 1.12 is tested.\n4. Clone the repo: `git clone petridishnn`\n5. Install dependency packages `python -m pip install -r \u003cpath_to_petridishnn\u003e/requirements.txt`\n6. Petridish needs some environment variables:\n`GLOBAL_LOG_DIR`: directory where logs will be written to by jobs running locally.\n`GLOBAL_MODEL_DIR`: directory where models will be written to by jobs running locally.\n`GLOBAL_DATA_DIR`: directory from where local jobs will read data.\nSet them to appropriate values in your bashrc. E.g. `export GLOBAL_MODEL_DIR=\"/home/dedey/data\"`\n\n## Getting the data\nPetridish code assumes datasets are in certain format (e.g. we transform ImageNet raw data to lmdb format).\nWhile one can always download the raw data of standard datasets and use the relevant scripts in `petridishnn/petridish/data` to convert\nthem Debadeepta Dey \u003cdedey@microsoft.com\u003e maintains an Azure blob with all the data in the converted format. (For Microsoft employees only)\nPlease email him for access.\n\n## Running a sample search job on cifar\nBefore doing full scale search on Azure it is common to check everything is running on local machine.\nAn example job script is at `petridishnn/scripts/test_distributed.sh`. Make sure you have all the\nenvironment variables used in this script. Run this from root folder of `petridishn` as `bash scripts/test_distributed.sh`.\nThis will output somethings to stdout but will output models and logs to the corresponding folders.\nIf this succeeds you have a working installation. Yay!\n\n## Post-search Analysis\n\nWe provide a number of scripts to analyze and post-process the search results in the directory\n[petridish/analysis](./petridish/analysis).\nWe also provide [a script to generate training scripts](./petridish/cust_exps_gen/generate_train_script.py) to train the found models.\nWe list them in the order of usage as follows.\nPlease refer to the header of each linked file for usage.\n\n1. [Inspect the search log](./petridish/analysis/search.py)\n2. [Generate scripts to train found models](./petridish/cust_exps_gen/generate_train_script.py)\n3. [Check performance of model training](./petridish/analysis/model.py)\n\n## Contacts:\n\n* Debadeepta Dey (dedey@microsoft.com)\n* Hanzhang Hu (hanzhang@cs.cmu.edu)\n* John Langford (jcl@microsoft.com)\n* Rich Caruana (rcaruana@microsoft.com)\n* Eric Horvitz (horvitz@microsoft.com)\n\n## Contributing\nPlease read the contributing [policy](./CONTRIBUTING.md) \n\n## Bibtex\nIf you would like to use this work for your research, please cite the following:\n```\n@article{hu2019forwardnas,\n  title={Efficient Forward Architecture Search},\n  author={Hanzhang Hu and John Langford and Rich Caruana and Saurajit Mukherjee and Eric Horvitz and Debadeepta Dey},\n  journal={Neural Information Processing Systems},\n  year={2019}\n}\n```\n","funding_links":[],"categories":["Uncategorized","Projects"],"sub_categories":["Uncategorized","Distributed Frameworks"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmicrosoft%2Fpetridishnn","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmicrosoft%2Fpetridishnn","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmicrosoft%2Fpetridishnn/lists"}