{"id":31737520,"url":"https://github.com/ucd4ids/monogenicfilterflux.jl","last_synced_at":"2025-10-09T09:22:54.102Z","repository":{"id":238576000,"uuid":"571850522","full_name":"UCD4IDS/MonogenicFilterFlux.jl","owner":"UCD4IDS","description":"An implementation of rotated monogenic scattering transform network","archived":false,"fork":false,"pushed_at":"2024-04-10T03:30:12.000Z","size":55,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2024-05-06T19:59:14.247Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Julia","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/UCD4IDS.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}},"created_at":"2022-11-29T02:33:33.000Z","updated_at":"2024-05-06T19:59:16.450Z","dependencies_parsed_at":"2024-05-06T20:09:19.507Z","dependency_job_id":null,"html_url":"https://github.com/UCD4IDS/MonogenicFilterFlux.jl","commit_stats":null,"previous_names":["ucd4ids/monogenicfilterflux.jl"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/UCD4IDS/MonogenicFilterFlux.jl","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/UCD4IDS%2FMonogenicFilterFlux.jl","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/UCD4IDS%2FMonogenicFilterFlux.jl/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/UCD4IDS%2FMonogenicFilterFlux.jl/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/UCD4IDS%2FMonogenicFilterFlux.jl/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/UCD4IDS","download_url":"https://codeload.github.com/UCD4IDS/MonogenicFilterFlux.jl/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/UCD4IDS%2FMonogenicFilterFlux.jl/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279001126,"owners_count":26083021,"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","status":"online","status_checked_at":"2025-10-09T02:00:07.460Z","response_time":59,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":[],"created_at":"2025-10-09T09:22:53.076Z","updated_at":"2025-10-09T09:22:54.096Z","avatar_url":"https://github.com/UCD4IDS.png","language":"Julia","funding_links":[],"categories":[],"sub_categories":[],"readme":"An implementation of rotated monogenic scattering transform network of 2-D signal $g$.\n\nLet $\\theta \\in [0, 2 \\pi)$. Denote the Riesz kernel $r_l$ and Riesz transform $R_l$ as  \n```math\nr_l({\\bf x}) = c_l \\dfrac{x_l}{\\Vert {\\bf x} \\Vert^3}\n```\nand\n```math\nR_l g({\\bf x}) = \\tilde{g}^{(l)}({\\bf x}) = (r_l ** g)({\\bf x}).\n```\n\nDenote the rotated signal as\n```math\n\\tilde{g}_{-\\theta}({\\bf x}) = g({\\bf R}_\\theta {\\bf x}).\n```\n\nThe Risez tranform of the rotated signal is\n```math\n\\tilde{g}^{(l)}_{-\\theta}({\\bf x}) = R_l \\tilde{g}_{-\\theta}({\\bf x}).\n```\n\nThe rotated monogenic decomposition is \n```math\n\\tilde{g}^{\\pm}_{-\\theta, \\theta}({\\bf x}) = g({\\bf x}) \\pm \\bigg( {\\bf i} \\tilde{g}^{(1)}_{-\\theta}({\\bf R}_{-\\theta} {\\bf x}) + {\\bf j} \\tilde{g}^{(2)}_{-\\theta}({\\bf R}_{-\\theta} {\\bf x} ) \\bigg).\n```\n\nBefore running the code, load packages\n```\nusing CUDA\nusing ScatteringTransform\nusing MonogenicFilterFlux\n```\n\nSuppose the feature dimension is `(nTrain_x, nTrain_y, 1, nSubsample)`.\n\nThe $l$-th layer monogenic scattering transform network with maximum scale $s$ is \n```\nscale = s; \nst = stFlux((nTrain_x, nTrain_y, 1, nSubsample), 2, σ=abs, outputPool = 1, scale = scale); \nst = cu(st);\n```\n\nSuppose `input_data` is the input of the 2-D signal, and set $\\theta$ as `ang`. \n\nWe can find the rotated monogenic decomposition of the scattering output by\n```\nimg_rot = rotate_image(input_data, ang);\noutput_rot = st(cu(img_rot))\noutput_rot = inv_rotate_out(output_rot, img_rot, -ang);\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fucd4ids%2Fmonogenicfilterflux.jl","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fucd4ids%2Fmonogenicfilterflux.jl","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fucd4ids%2Fmonogenicfilterflux.jl/lists"}