{"id":20669545,"url":"https://github.com/iitis/autoencodertestingenvironment","last_synced_at":"2025-07-23T09:37:01.274Z","repository":{"id":141460255,"uuid":"379221962","full_name":"iitis/AutoencoderTestingEnvironment","owner":"iitis","description":"Experimental and testing environment for autoencoders","archived":false,"fork":false,"pushed_at":"2021-06-30T12:11:01.000Z","size":105,"stargazers_count":2,"open_issues_count":0,"forks_count":2,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-03-17T17:59:35.171Z","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/iitis.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,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-06-22T09:59:31.000Z","updated_at":"2023-06-20T10:12:27.000Z","dependencies_parsed_at":null,"dependency_job_id":"c607ecc3-8e5a-4f2f-a1b3-d9a7f58b2441","html_url":"https://github.com/iitis/AutoencoderTestingEnvironment","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/iitis/AutoencoderTestingEnvironment","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iitis%2FAutoencoderTestingEnvironment","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iitis%2FAutoencoderTestingEnvironment/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iitis%2FAutoencoderTestingEnvironment/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iitis%2FAutoencoderTestingEnvironment/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/iitis","download_url":"https://codeload.github.com/iitis/AutoencoderTestingEnvironment/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iitis%2FAutoencoderTestingEnvironment/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266655747,"owners_count":23963554,"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":"2025-07-23T02:00:09.312Z","response_time":66,"last_error":null,"robots_txt_status":null,"robots_txt_updated_at":null,"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":[],"created_at":"2024-11-16T20:14:45.919Z","updated_at":"2025-07-23T09:37:01.231Z","avatar_url":"https://github.com/iitis.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"## DESCRIPTION:\n\nAutoencoder Testing Environment (ATE) v.1.0\nExperimental and testing environment for autoencoders\n\nRelated to the work:\n\u003e Stable training of autoencoders for hyperspectral unmixing\n\nSource code for the review process of the 28th International Conference on Neural Information Processing (ICONIP 2021).\n\n## LICENSE:\nCopyright 2021 Institute of Theoretical and Applied Informatics,\nPolish Academy of Sciences (ITAI PAS) \u003chttps://www.iitis.pl\u003e\nAuthors:\n- Kamil Książek (ITAI PAS, ORCID ID: [0000−0002−0201−6220](https://orcid.org/0000-0002-0201-6220)),\n- Przemysław Głomb (ITAI PAS, ORCID ID: [0000−0002−0215−4674](https://orcid.org/0000-0002-0215-4674)),\n- Michał Romaszewski (ITAI PAS, ORCID ID: [0000−0002−8227−929X](https://orcid.org/0000-0002-8227-929X)),\n- Michał Cholewa (ITAI PAS, ORCID ID: [0000−0001−6549−1590](https://orcid.org/0000-0001-6549-1590)),\n- Bartosz Grabowski (ITAI PAS, ORCID ID: [0000−0002−2364−6547](https://orcid.org/0000-0002-2364-6547)).\n\nThis program is free software: you can redistribute it and/or modify\nit under the terms of the GNU General Public License as published by\nthe Free Software Foundation, either version 3 of the License, or\n(at your option) any later version.\n\nThis program is distributed in the hope that it will be useful,\nbut WITHOUT ANY WARRANTY; without even the implied warranty of\nMERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\nGNU General Public License for more details.\n\nYou should have received a copy of the GNU General Public License\nalong with this program. If not, see \u003chttps://www.gnu.org/licenses/\u003e.\n\n## FUNCTIONALITY:\n- Autoencoder training \u0026 retraining in order to avoid bad initialisations described in the paper\n- Unmixing and reconstruction error/spectra evaluation\n- Simplex evaluation\n- RayTune hyperparameter selection (GS+ASHA)\n\n## RULES:\n\n- Autoencoders architectures are loaded from \u003cem\u003e./architectures/\u003c/em\u003e subfolder. One file per autoencoder.\n- An example experiment using a synthetic dataset can be run from `exp_demo.py` file.\n\n## FILES:\n\n- `ate/ate_*.py`: Autoencoder Testing Environment core files.\n- `ate/ate_tests*.py`: ATE tests.\n- `exp_*.py`: Experiment files.\n- `architectures/*.py`: Autoencoder files. One file contains one autoencoder.\n- `util_*.py`: External libraries (to be removed in future).\n\n## DATASETS:\n\nAll datasets have to be inserted into \u003cem\u003e./data/\u003c/em\u003e folder. The example, synthetic \u003cem\u003eCustom\u003c/em\u003e  is used by the demonstration experiment file (`exp_demo.py`).\n\n## ARCHITECTURES:\n\nThree autoencoder architectures are prepared:\n\n- `original`: an architecture with sigmoid activation function from [Palsson et al.](https://ieeexplore.ieee.org/document/8322133) paper;\n- `modified`: a modification of the above architecture;\n- `basic`: a simple architecture with ReLU activation function.\n\n## USAGE:\n\nCopy and rename `exp_demo.py` file to `exp_\u003cyour_name_here\u003e.py` file.\nEnsure that paths in params_globals, params_aa are correct (see demo).\nPut your autoencoders into \u003cem\u003earchitectures/\u003c/em\u003e directory with the \u003cem\u003eAutoencoder\u003c/em\u003e class name.\nTests use parameters in `ate_tests.tests.params.py`, ensure that paths are consistent with paths in params.\n\n\n## TUNE USAGE:\n\n1. Copy and rename `exp_demo.py` file to `exp_\u003cyour_name_here\u003e.py` file.\n2. Put your autoencoders into \u003cem\u003earchitectures/\u003c/em\u003e directory with the \u003cem\u003eAutoencoder\u003c/em\u003e class name.\n3. Modify `run()` function's body:\n    1. The only hyperparameter from `default_params_aa` that is used is the `no_epochs`, which describes for how many epochs the model will be trained after finding the best parameters.\n    2. You need to define\n      \t- `autoencoder_name` (name that is recognizable by `get_autoencoder`, e.g. \u003cem\u003e'basic'\u003c/em\u003e);\n      \t- `dataset_name` (name recognizable by `get_dataset`, e.g. \u003cem\u003e'Custom'\u003c/em\u003e);\n      \t- `params_aa`, `params_global` - dictionaries with hyperparameters described in \u003cem\u003e'exp_demo.py\u003c/em\u003e;\n\t\t- `tune_config`: parameters with their ranges to be found by the RayTune optimizer, details are contained in the architecture files;\n\t\t- `loss_function`: loss function for optimization, possible options: Mean Squared Error `LossMSE()` or Spectral Angle Distance `LossSAD()`;\n\t\t- `experiment_name`: anything which can be a filename;\n\t\t- `grace`: parameter of ASHA Scheduler defining interval between stopping of trials;\n\t\t- `no_epochs`: the maximum number of epochs during RayTune optimization;\n\t\t- `num_samples`: parameter defining a number of sampling from the space of hyperparameters;\n\t\t- `resources_per_trial`: CPU and GPU resources to allocate per trial.\n\t\tMore details connected with RayTune parameter are contained in [the official RayTune documentation](https://docs.ray.io/en/stable/index.html).\n    3. invoke `experiment_tune()` function passing arguments as described in its definition and docstring.\n4. In `if __name__ == \"__main__\"`'s body\n   1. init_env()\n   2. run()\n5. Run the file from shell with `python exp_\u003cyour_name\u003e.py`.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fiitis%2Fautoencodertestingenvironment","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fiitis%2Fautoencodertestingenvironment","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fiitis%2Fautoencodertestingenvironment/lists"}