{"id":20677738,"url":"https://github.com/guildai/rare-event","last_synced_at":"2026-04-22T19:33:00.749Z","repository":{"id":110397847,"uuid":"200533492","full_name":"guildai/rare-event","owner":"guildai","description":null,"archived":false,"fork":false,"pushed_at":"2019-08-05T19:26:27.000Z","size":4044,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-03-10T19:25:25.650Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/guildai.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2019-08-04T19:30:46.000Z","updated_at":"2019-08-05T19:26:29.000Z","dependencies_parsed_at":null,"dependency_job_id":"c14db8d2-9a35-44e3-8f0f-b1358122e95b","html_url":"https://github.com/guildai/rare-event","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/guildai/rare-event","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/guildai%2Frare-event","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/guildai%2Frare-event/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/guildai%2Frare-event/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/guildai%2Frare-event/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/guildai","download_url":"https://codeload.github.com/guildai/rare-event/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/guildai%2Frare-event/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32152592,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-22T17:06:48.269Z","status":"ssl_error","status_checked_at":"2026-04-22T17:06:19.037Z","response_time":58,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":"2024-11-16T21:16:54.557Z","updated_at":"2026-04-22T19:33:00.715Z","avatar_url":"https://github.com/guildai.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Rare Event Prediction\n\n[These models](guild.yml) are adapted from these blog posts:\n\n- [Extreme Rare Event Classification using Autoencoders in Keras](https://towardsdatascience.com/extreme-rare-event-classification-using-autoencoders-in-keras-a565b386f098)\n- [LSTM Autoencoder for Extreme Rare Event Classification in Keras](https://towardsdatascience.com/lstm-autoencoder-for-extreme-rare-event-classification-in-keras-ce209a224cfb)\n\nby [Chitta Ranjan](https://www.linkedin.com/in/chitta-ranjan-b0851911/)\n\nThe original source is included as Notebooks:\n\n- [autoencoder_classifier.ipynb](autoencoder_classifier.ipynb)\n- [lstm_autoencoder_classifier.ipynb](lstm_autoencoder_classifier.ipynb)\n\nTo train the models, use:\n\n    $ guild run ae:train\n    $ guild run lstm:train\n\nThe LSTM does not include validation accuracy.\n\n## To Do\n\n- [ ] Generate sample log (treat as simulation problem)\n  - Contains mostly normal log events of whatever (negative example)\n  - Supports SIGTERM or some other signal\n  - Prints signal\n  - After some period with a random component, logs a \"crash\"\n    (positive example)\n- [ ] Convert simulated logs into format we can train\n- [x] Activation functions (elu, leaky relu, etc) (see advanced\n      activations in Keras)\n- [x] More or fewer layers\n- [ ] Different optimizers\n- [ ] Within the LSTM:\n  - Dropout\n  - ???\n- [x] Bump epochs to 1000\n- [x] Add early stopping (Keras callback)\n- [ ] Learning rate schedules\n- [ ] Use custom Keras metic for roc_auc (unless slows training)\n- [ ] Check if metrics for LSTM is slowing training\n\n### Bug in data processing\n\n- Losing a column somehow\n- He's using the row number in the xs, which masks the missing col\n\n------------------\n\n- Highlight feature engineering in data-preparation (convert from raw\n  to prepared - time shift of y values)\n\n- Use validation data for examples\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fguildai%2Frare-event","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fguildai%2Frare-event","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fguildai%2Frare-event/lists"}