{"id":13499097,"url":"https://github.com/titu1994/Keras-DualPathNetworks","last_synced_at":"2025-03-29T03:32:20.626Z","repository":{"id":150946819,"uuid":"99740775","full_name":"titu1994/Keras-DualPathNetworks","owner":"titu1994","description":"Dual Path Networks for Keras 2.0+","archived":false,"fork":false,"pushed_at":"2020-02-09T15:54:15.000Z","size":427,"stargazers_count":113,"open_issues_count":3,"forks_count":52,"subscribers_count":9,"default_branch":"master","last_synced_at":"2025-03-16T08:41:17.246Z","etag":null,"topics":["deep-learning","dpn","dual-path-networks","keras","tensorflow"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","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,"governance":null}},"created_at":"2017-08-08T22:17:33.000Z","updated_at":"2025-01-23T05:53:54.000Z","dependencies_parsed_at":null,"dependency_job_id":"3e23fcf1-9df0-46e4-b96d-635ca1911274","html_url":"https://github.com/titu1994/Keras-DualPathNetworks","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/titu1994%2FKeras-DualPathNetworks","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/titu1994%2FKeras-DualPathNetworks/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/titu1994%2FKeras-DualPathNetworks/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/titu1994%2FKeras-DualPathNetworks/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/titu1994","download_url":"https://codeload.github.com/titu1994/Keras-DualPathNetworks/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246135765,"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":["deep-learning","dpn","dual-path-networks","keras","tensorflow"],"created_at":"2024-07-31T22:00:28.583Z","updated_at":"2025-03-29T03:32:20.358Z","avatar_url":"https://github.com/titu1994.png","language":"Python","funding_links":[],"categories":["Papers\u0026Codes"],"sub_categories":["DPN"],"readme":"# Dual Path Networks in Keras\n[Dual Path Networks](https://arxiv.org/abs/1707.01629) are highly efficient networks which combine the strength of both ResNeXt [Aggregated Residual Transformations for Deep Neural Networks](https://arxiv.org/abs/1611.05431) and DenseNets [Densely Connected Convolutional Networks](https://arxiv.org/abs/1608.06993).\n\nNote: Weights have not been ported over yet.\n\n## Dual Path Connections\n\u003cimg src=\"https://github.com/titu1994/Keras-DualPathNetworks/blob/master/images/dual%20path%20networks.png?raw=true\" width=\"100%\" height=\"100%\"\u003e\n\n## Usage\nSeveral of the standard Dual Path Network models have been included. They can be initialized as : \n```\nfrom dual_path_network import DPN92, DPN98, DPN107, DPN137\n\nmodel = DPN92(input_shape=(224, 224, 3)) # same for the others\n```\n\nTo create a custom DualPathNetwork, use the DualPathNetwork builder method : \n```\nfrom dual_path_network import DualPathNetwork\n\nmodel = DualPathNetwork(input_shape=(224, 224, 3),\n                        initial_conv_filters=64,\n                        depth=[3, 4, 20, 3],\n                        filter_increment=[16, 32, 24, 128],\n                        cardinality=32,\n                        width=3,\n                        weight_decay=0,\n                        include_top=True,\n                        weights=None,\n                        input_tensor=None,\n                        pooling=None,\n                        classes=1000)\n```\n\n## Performance\n\u003cimg src=\"https://github.com/titu1994/Keras-DualPathNetworks/blob/master/images/original-results-on-imagenet1k.png?raw=true\" height=100% width=100%\u003e\n\n## Support \n- Keras does not have inbuilt support for grouped convolutions. Therefore I had to use lambda layers to match the ResNeXt paper implementation. When grouped convolution support is added, I hope to add it in this as well.\n- Mean-Max Global Pooling support is present with the help of Lambda layer to scale the sum.\n- Depth and Filter_Increment must be lists for now, and must be lists of same length. Will think about adding support for integers, but I think list support is far more useful anyway, so I may not implement it.\n- Weight decay support is added, but disabled by default. The DPN paper does not mention it, but ResNet, WRN and ResNeXt paper may all use small weight regularization. Use a small value of `1e-4` or `5e-4` if you wish to use it.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftitu1994%2FKeras-DualPathNetworks","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftitu1994%2FKeras-DualPathNetworks","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftitu1994%2FKeras-DualPathNetworks/lists"}