{"id":13738509,"url":"https://github.com/Erlemar/pytorch_tempest","last_synced_at":"2025-05-08T16:34:20.842Z","repository":{"id":46070359,"uuid":"291450945","full_name":"Erlemar/pytorch_tempest","owner":"Erlemar","description":"My repo for training neural nets using pytorch-lightning and hydra","archived":false,"fork":false,"pushed_at":"2024-08-03T07:27:14.000Z","size":119483,"stargazers_count":208,"open_issues_count":5,"forks_count":21,"subscribers_count":8,"default_branch":"master","last_synced_at":"2024-08-04T03:12:43.901Z","etag":null,"topics":["deep-learning","hacktoberfest","hydra","pytorch-lightning","training-pipeline"],"latest_commit_sha":null,"homepage":"https://pytorch-tempest.readthedocs.io/en/latest/","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/Erlemar.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"contributing.md","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}},"created_at":"2020-08-30T10:46:59.000Z","updated_at":"2024-08-03T07:27:16.000Z","dependencies_parsed_at":"2024-04-16T22:50:50.906Z","dependency_job_id":"3cf53202-1846-4658-aca8-c69c78d2294a","html_url":"https://github.com/Erlemar/pytorch_tempest","commit_stats":null,"previous_names":[],"tags_count":0,"template":true,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Erlemar%2Fpytorch_tempest","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Erlemar%2Fpytorch_tempest/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Erlemar%2Fpytorch_tempest/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Erlemar%2Fpytorch_tempest/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Erlemar","download_url":"https://codeload.github.com/Erlemar/pytorch_tempest/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224746824,"owners_count":17363123,"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":["deep-learning","hacktoberfest","hydra","pytorch-lightning","training-pipeline"],"created_at":"2024-08-03T03:02:24.515Z","updated_at":"2024-11-15T07:31:17.450Z","avatar_url":"https://github.com/Erlemar.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# tempest\n\n[![DeepSource](https://static.deepsource.io/deepsource-badge-light-mini.svg)](https://deepsource.io/gh/Erlemar/pytorch_tempest/?ref=repository-badge)\n\nThis repository has my pipeline for training neural nets.\n\nMain frameworks used:\n\n* [hydra](https://github.com/facebookresearch/hydra)\n* [pytorch-lightning](https://github.com/PyTorchLightning/pytorch-lightning)\n\nThe main ideas of the pipeline:\n\n* all parameters and modules are defined in configs;\n* prepare configs beforehand for different optimizers/schedulers and so on, so it is easy to switch between them;\n* have templates for different deep learning tasks. Currently, image classification and named entity recognition are supported;\n\nExamples of running the pipeline:\nThis will run training on MNIST (data will be downloaded):\n```shell\n\u003e\u003e\u003e python train.py --config-name mnist_config model.encoder.params.to_one_channel=True\n```\n\nRunning on MPS (M1 macbook)\n```shell\npython train.py --config-name mnist_config model.encoder.params.to_one_channel=True trainer.accelerator=mps +trainer.devices=1 optimizer=adan training.lr=0.001\n```\nRunning on MPS (M1 macbook) with schedule free optimizer https://github.com/facebookresearch/schedule_free/tree/main\n```shell\npython train.py --config-name mnist_config model.encoder.params.to_one_channel=True trainer.accelerator=mps trainer.devices=1 optimizer=adamwschedulefree training.lr=0.001 scheduler.params.patience=100\n\n```\nThe default run:\n\n```shell\n\u003e\u003e\u003e python train.py\n```\n\nThe default version of the pipeline is run on imagenette dataset. To do it, download the data from this repository:\nhttps://github.com/fastai/imagenette\nunzip it and define the path to it in conf/datamodule/image_classification.yaml path\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FErlemar%2Fpytorch_tempest","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FErlemar%2Fpytorch_tempest","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FErlemar%2Fpytorch_tempest/lists"}