{"id":19898937,"url":"https://github.com/johnolafenwa/torchfusion-utils","last_synced_at":"2025-08-01T07:31:32.367Z","repository":{"id":62584971,"uuid":"184654739","full_name":"johnolafenwa/TorchFusion-Utils","owner":"johnolafenwa","description":"A pytorch helper library for Mixed Precision Training, Initialization, Metrics and More Utilities to simplify training of deep learning models","archived":false,"fork":false,"pushed_at":"2019-06-08T11:02:00.000Z","size":63,"stargazers_count":15,"open_issues_count":0,"forks_count":7,"subscribers_count":2,"default_branch":"master","last_synced_at":"2024-12-05T11:05:55.221Z","etag":null,"topics":["deeplearning","initializer","machine-learning","metrics","mixed-precision-training","nvidia","python","pytorch"],"latest_commit_sha":null,"homepage":"https://utils.torchfusion.org","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/johnolafenwa.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}},"created_at":"2019-05-02T21:36:45.000Z","updated_at":"2023-04-28T13:04:56.000Z","dependencies_parsed_at":"2022-11-03T22:03:09.631Z","dependency_job_id":null,"html_url":"https://github.com/johnolafenwa/TorchFusion-Utils","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/johnolafenwa%2FTorchFusion-Utils","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/johnolafenwa%2FTorchFusion-Utils/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/johnolafenwa%2FTorchFusion-Utils/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/johnolafenwa%2FTorchFusion-Utils/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/johnolafenwa","download_url":"https://codeload.github.com/johnolafenwa/TorchFusion-Utils/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":228348393,"owners_count":17905898,"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":["deeplearning","initializer","machine-learning","metrics","mixed-precision-training","nvidia","python","pytorch"],"created_at":"2024-11-12T20:06:10.842Z","updated_at":"2024-12-05T17:58:17.679Z","avatar_url":"https://github.com/johnolafenwa.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# TorchFusion-Utils\nA pytorch helper library for Mixed Precision Training, Metrics and More Utilities to simplify training of deep learning models.\n\nTorchFusion Utils was built to enable pytorch programmers easily take advantage of advanced training techniques without having to use any specific trainer framework. It is very transparent and can be easily plugged in to existing [Pytorch](https://pytorch.org) code bases\n\n# Installation\n\nTorchFusion Utils is extremely light with no other dependency other than [Pytorch](https://pytorch.org) itself. \nYou can install from pypi\n\n\u003cpre\u003e pip3 install torchfusion-utils --upgrade \u003c/pre\u003e\n\n\n# Core Features\n\n**Mixed Precision Training**\n\nIn just two lines of code, you can speed up training of your [Pytorch](https://pytorch.org) models, reduce memory usage on any GPU and fit in larger batch sizes than was previously possible on your GPU.\n\n\u003cpre\u003e#convert your model and optimizer to mixed precision mode\nmodel, optim = convertToFP16(model,optim)\n\n#in your batch loop, replace loss.backward with optim.backward(loss)\noptim.backward(loss)\n\u003c/pre\u003e\n\n**Initialization**\n\nA very simple api to easily initialize your model parameters with fine grained control over the type of layers and type of weights to be initialized.\n\n\u003cpre\u003e\nkaiming_normal_init(model,types=[nn.Conv2d],category=\"weight\")\n\u003c/pre\u003e\n\n**Metrics**\n\nAn extensible metric package that makes it easy to easily compute accuracy of your models. A number of metrics are provided out of the box and you can extend to add yours.\n\n\u003cpre\u003e\ntop5_acc = Accuracy(topk=5)\n\n#sample evaluation loop\nfor i,(x,y) in enumerate(data_loader):\n    predictions = model(x)\n    top5_acc.update(predictions,y)\n\nprint(\"Top 5 Acc: \",top5_accc.getValue())\n\u003c/pre\u003e\n\n**Model Utilities**\n\nSimple functions to easily analyse, load and save your [Pytorch](https://pytorch.org) models in an error free way.\n\n# Documentation\n\nFind tutorials and extensive documentation on using TorchFusion Utils from [https://utils.torchfusion.org](https://utils.torchfusion.org)\n\n\n# About The TorchFusion Project\n\nThe TorchFusion project is a set of [Pytorch](https://pytorch.org) based deep learning libraries aimed at making making research easier and more productive. We believe anyone can be a great researcher with the right tools, thats why we build!\n\nTorchFusion is an initiative of [DeepQuest AI](https://deepquestai.com), founded by John Olafenwa \u0026 Moses Olafenwa.\n\n\u003c/pre\u003e\n\n\n\u003ch3\u003e\u003cb\u003e\u003cu\u003eContact Developers\u003c/u\u003e\u003c/b\u003e\u003c/h3\u003e\n \u003cp\u003e\n  \u003cbr\u003e\n      \u003cb\u003eJohn Olafenwa\u003c/b\u003e \u003cbr\u003e\n    \u003ci\u003eEmail: \u003c/i\u003e    \u003ca style=\"text-decoration: none;\"  href=\"mailto:johnolafenwa@gmail.com\"\u003e johnolafenwa@gmail.com\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eWebsite: \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" target=\"_blank\" href=\"https://john.aicommons.science\"\u003e https://john.aicommons.science\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eTwitter: \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" target=\"_blank\" href=\"https://twitter.com/johnolafenwa\"\u003e @johnolafenwa\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eMedium : \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" target=\"_blank\" href=\"https://medium.com/@johnolafenwa\"\u003e @johnolafenwa\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eFacebook : \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" href=\"https://facebook.com/olafenwajohn\"\u003e olafenwajohn\u003c/a\u003e \u003cbr\u003e\n\n\u003cbr\u003e\n  \u003cb\u003eMoses Olafenwa\u003c/b\u003e \u003cbr\u003e\n    \u003ci\u003eEmail: \u003c/i\u003e    \u003ca style=\"text-decoration: none;\"  href=\"mailto:guymodscientist@gmail.com\"\u003e guymodscientist@gmail.com\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eWebsite: \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" target=\"_blank\" href=\"https://moses.aicommons.science\"\u003e https://moses.aicommons.science\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eTwitter: \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" target=\"_blank\" href=\"https://twitter.com/OlafenwaMoses\"\u003e @OlafenwaMoses\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eMedium : \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" target=\"_blank\" href=\"https://medium.com/@guymodscientist\"\u003e @guymodscientist\u003c/a\u003e \u003cbr\u003e\n      \u003ci\u003eFacebook : \u003c/i\u003e    \u003ca style=\"text-decoration: none;\" target=\"_blank\" href=\"https://facebook.com/moses.olafenwa\"\u003e moses.olafenwa\u003c/a\u003e \u003cbr\u003e\n\u003cbr\u003e\n \u003c/p\u003e\n\n \u003cbr\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjohnolafenwa%2Ftorchfusion-utils","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjohnolafenwa%2Ftorchfusion-utils","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjohnolafenwa%2Ftorchfusion-utils/lists"}