{"id":28421819,"url":"https://github.com/viktor-shcherb/triage","last_synced_at":"2026-02-12T01:38:44.706Z","repository":{"id":45677823,"uuid":"514014986","full_name":"viktor-shcherb/triage","owner":"viktor-shcherb","description":"Script running tool for optimizing GPU memory utilization","archived":false,"fork":false,"pushed_at":"2025-05-02T11:33:23.000Z","size":44,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-09-20T11:45:39.753Z","etag":null,"topics":["automation","cli","cuda","deep-learning","devops-tools","experiment-runner","gpu-monitoring","gpu-scheduler","hyperparameter-sweep","job-queue","machine-learning","nvidia-smi","pypi-package","python","resource-management","script-runner"],"latest_commit_sha":null,"homepage":"","language":"Python","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/viktor-shcherb.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,"zenodo":null}},"created_at":"2022-07-14T18:54:36.000Z","updated_at":"2025-07-28T20:23:37.000Z","dependencies_parsed_at":"2025-06-05T07:16:33.388Z","dependency_job_id":"b931af4a-907b-4ed8-9ae8-905129c994b2","html_url":"https://github.com/viktor-shcherb/triage","commit_stats":null,"previous_names":["viktor-shcherb/triage","viktorooreps/triage"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/viktor-shcherb/triage","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/viktor-shcherb%2Ftriage","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/viktor-shcherb%2Ftriage/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/viktor-shcherb%2Ftriage/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/viktor-shcherb%2Ftriage/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/viktor-shcherb","download_url":"https://codeload.github.com/viktor-shcherb/triage/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/viktor-shcherb%2Ftriage/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29352953,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-12T01:03:07.613Z","status":"ssl_error","status_checked_at":"2026-02-12T01:00:51.346Z","response_time":97,"last_error":"SSL_read: 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":["automation","cli","cuda","deep-learning","devops-tools","experiment-runner","gpu-monitoring","gpu-scheduler","hyperparameter-sweep","job-queue","machine-learning","nvidia-smi","pypi-package","python","resource-management","script-runner"],"created_at":"2025-06-05T07:10:55.017Z","updated_at":"2026-02-12T01:38:44.680Z","avatar_url":"https://github.com/viktor-shcherb.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# TRIAGE\n\nTriage is a lightweight Python runner that turns a pile of training scripts into a self-managed queue, launching each job only when the GPUs have enough free memory. Give it one or more JSON “run configurations” with a memory budget and argument grid; Triage watches nvidia-smi, starts the job the moment resources are free, iterates through all parameter combinations, and records the run under a unique task name. It keeps shared servers from low VRAM utilization and lets you squeeze every last gigabyte out of your hardware during deep-learning experiments.\n\n## Installation\n\n```bash\npip install triage-runner\n```\n\n## Usage\n\nSee `--help` option for extended list of possible arguments.  \nRunning one config:\n```bash\ntriage run_config.json\n```\nRunning several configs:\n```bash\ntriage run_config1.json run_config2.json run_config3.json \n```\nPatterns can be used for config discovery as well:\n```bash\ntriage run_config*.json\n```\nMore on pattern syntax can be found here: https://docs.python.org/3.10/library/pathlib.html#pathlib.Path.glob\n\n## Run configurations\n\nStored in JSON format. The sample run configuration looks like this:\n```json\n{\n  \"memory_needed\": 10.0,\n  \"config_name\": \"sample_config\",\n  \"command\": \"python3 train.py\",\n  \"args\": [\n    \"arg1\",\n    \"--arg2\",\n    [\"--seed=1\", \"--seed=2\", \"--seed=3\"],\n    \"--arg3=3\"\n  ]\n}\n```\nEvery entry in `args` list is an argument for `command`. An entry can be a list - in which case TRIAGE will iterate through all the possible combinations of all values in list entries. The example script above will be run 3 times with an argument `--seed` set to 1, 2 and 3.\n\nParameter `config_name` is optional and is used for logging the results (see `--logfile` option). Based on this parameter environment variable `TASK_NAME` is set in order to be used by running script.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fviktor-shcherb%2Ftriage","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fviktor-shcherb%2Ftriage","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fviktor-shcherb%2Ftriage/lists"}