{"id":13499076,"url":"https://github.com/titu1994/Keras-ResNeXt","last_synced_at":"2025-03-29T03:32:17.593Z","repository":{"id":52833013,"uuid":"93877235","full_name":"titu1994/Keras-ResNeXt","owner":"titu1994","description":"Implementation of ResNeXt models from the paper Aggregated Residual Transformations for Deep Neural Networks in Keras 2.0+.","archived":false,"fork":false,"pushed_at":"2021-04-17T10:05:32.000Z","size":645,"stargazers_count":224,"open_issues_count":8,"forks_count":79,"subscribers_count":8,"default_branch":"master","last_synced_at":"2025-03-24T08:05:38.300Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","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/titu1994.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":"2017-06-09T16:25:11.000Z","updated_at":"2024-10-21T13:34:01.000Z","dependencies_parsed_at":"2022-08-22T22:40:32.332Z","dependency_job_id":null,"html_url":"https://github.com/titu1994/Keras-ResNeXt","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/titu1994%2FKeras-ResNeXt","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/titu1994%2FKeras-ResNeXt/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/titu1994%2FKeras-ResNeXt/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/titu1994%2FKeras-ResNeXt/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/titu1994","download_url":"https://codeload.github.com/titu1994/Keras-ResNeXt/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246135766,"owners_count":20729056,"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":[],"created_at":"2024-07-31T22:00:28.093Z","updated_at":"2025-03-29T03:32:17.569Z","avatar_url":"https://github.com/titu1994.png","language":"Python","funding_links":[],"categories":["Papers\u0026Codes"],"sub_categories":["ResNext"],"readme":"# Keras ResNeXt\n\nImplementation of ResNeXt models from the paper [Aggregated Residual Transformations for Deep Neural Networks](https://arxiv.org/pdf/1611.05431.pdf) in Keras 2.0+.\n\nContains code for building the general ResNeXt model (optimized for datasets similar to CIFAR) and ResNeXtImageNet (optimized for the ImageNet dataset).\n\n# Salient Features\nResNeXt updates the ResNet block with a new expanded block architecture, which depends on the `cardinality` parameter. It can be further visualised in the below diagram from the paper.\n\n![cardinality](https://github.com/titu1994/Keras-ResNeXt/blob/master/images/Cardinality.PNG?raw=true)\n\n---\nHowever, since grouped convolutions are not directly available in Keras, an equivalent variant is used in this repository (see block 2)\n\n![equivalent blocks](https://github.com/titu1994/Keras-ResNeXt/blob/master/images/equivalent_blocks.PNG?raw=true)\n\n# Usage\n\nFor the general ResNeXt model (for all datasets other than ImageNet),\n\n```\nfrom resnext import ResNext\n\nmodel = ResNext(image_shape, depth, cardinality, width, weight_decay)\n```\n\nFor the ResNeXt model which has been optimized for ImageNet,\n\n```\nfrom resnext import ResNextImageNet\n\nimage_shape = (112, 112, 3) if K.image_data_format() == 'channels_last' else (3, 112, 112)\nmodel = ResNextImageNet(image_shape)\n```\n\nNote, there are other parameters such as depth, cardinality, width and weight_decay just as in the general model, however the defaults are set according to the paper.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftitu1994%2FKeras-ResNeXt","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftitu1994%2FKeras-ResNeXt","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftitu1994%2FKeras-ResNeXt/lists"}