{"id":20360760,"url":"https://github.com/masadcv/propel","last_synced_at":"2025-07-22T10:33:04.823Z","repository":{"id":57476560,"uuid":"341742750","full_name":"masadcv/PROPEL","owner":"masadcv","description":"Official PyTorch implementation for \"PROPEL: Probabilistic Parametric Regression Loss for Convolutional Neural Networks\"","archived":false,"fork":false,"pushed_at":"2022-03-15T03:11:26.000Z","size":2872,"stargazers_count":6,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-06-17T03:26:58.255Z","etag":null,"topics":["deep-learning","loss-function","machine-learning","probabilistic","pytorch-implementation","regression"],"latest_commit_sha":null,"homepage":"https://masadcv.github.io/PROPEL/index.html","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/masadcv.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":"2021-02-24T01:32:38.000Z","updated_at":"2022-07-13T17:58:44.000Z","dependencies_parsed_at":"2022-09-12T15:22:33.969Z","dependency_job_id":null,"html_url":"https://github.com/masadcv/PROPEL","commit_stats":null,"previous_names":[],"tags_count":2,"template":false,"template_full_name":null,"purl":"pkg:github/masadcv/PROPEL","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/masadcv%2FPROPEL","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/masadcv%2FPROPEL/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/masadcv%2FPROPEL/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/masadcv%2FPROPEL/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/masadcv","download_url":"https://codeload.github.com/masadcv/PROPEL/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/masadcv%2FPROPEL/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266475602,"owners_count":23935000,"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":"2025-07-22T02:00:09.085Z","response_time":66,"last_error":null,"robots_txt_status":null,"robots_txt_updated_at":null,"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":["deep-learning","loss-function","machine-learning","probabilistic","pytorch-implementation","regression"],"created_at":"2024-11-14T23:42:52.459Z","updated_at":"2025-07-22T10:33:04.795Z","avatar_url":"https://github.com/masadcv.png","language":"Python","readme":"# PRObablistic Parametric rEgression Loss (PROPEL) \nPRObabilistic Parametric rEgresison Loss (PROPEL) is a loss function that enables probabilisitic regression for a neural network. It achieves this by enabling a neural network to learn parameters of a mixture of Gaussian distribution. \n\nFurther details about the loss can be found in the paper: [PROPEL: Probabilistic Parametric Regression Loss for Convolutional Neural Networks](https://arxiv.org/pdf/1807.10937.pdf)\n\nThis repository provides official pytorch implementation of PROPEL. \n\n# Installation Instructions\nPROPEL can be installed using the following command \n\n```bash\npip install torchpropel\n```\n\n\n```bash\npip install git+https://github.com/masadcv/PROPEL.git\n```\n\n# Usage Example\n```python\nimport torch\nimport numpy as np\n\nfrom torchpropel import PROPEL\n\n# Our example has a neural network with\n# output [num_batch, num_gaussians, num_dims]\nnum_batch = 4\nnum_gaussians = 6\nnum_dims = 3\n\n# setting ground-truth variance sigma_gt=0.2\nsigma_gt = 0.2\npropel_loss = PROPEL(sigma_gt)\n\n# ground truth targets for loss\ny = torch.ones((num_batch, num_dims)) * 0.5\n\n# example prediction - this can also be coming as output of a neural network\nfeat_g = np.random.randn(num_batch, num_gaussians, 2 * num_dims) * 0.5\nfeat_g[:, :, num_dims::] = 0.2\nfeat = torch.tensor(feat_g, dtype=y.dtype)\n\n# compute the loss\nL = propel_loss(feat, y)\n\nprint(L)\n```\n# Documentation\nFurther details of each function implemented for PROPEL can be accessed at the documentation hosted at: [https://masadcv.github.io/PROPEL/index.html](https://masadcv.github.io/PROPEL/index.html). \n\n# Citing PROPEL\nPre-print of PROPEL can be found at: [PROPEL: Probabilistic Parametric Regression Loss for Convolutional Neural Networks](https://arxiv.org/pdf/1807.10937.pdf)\n\nIf you use PROPEL in your research, then please cite:\n\nBibTeX:\n```\n@inproceedings{asad2020propel,\n  title={PROPEL: Probabilistic Parametric Regression Loss for Convolutional Neural Networks},\n  author={Asad, Muhammad and Basaru, Rilwan and Arif, SM and Slabaugh, Greg},\n  booktitle={25th International Conference on Pattern Recognition},\n  pages={},\n  year={2020}}\n```\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmasadcv%2Fpropel","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmasadcv%2Fpropel","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmasadcv%2Fpropel/lists"}