{"id":24847891,"url":"https://github.com/skent259/ordinal-mil-nnets","last_synced_at":"2025-08-22T13:11:39.355Z","repository":{"id":191542959,"uuid":"513343876","full_name":"skent259/ordinal-mil-nnets","owner":"skent259","description":"Experiments testing ordinal and multiple instance learning neural networks","archived":false,"fork":false,"pushed_at":"2023-08-31T03:38:51.000Z","size":31491,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-31T11:47:12.400Z","etag":null,"topics":["multiple-instance-learning","neural-network","ordinal","python","reproducible-research","simulation","weakly-supervised-learning"],"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/skent259.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,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2022-07-13T01:37:20.000Z","updated_at":"2023-08-31T03:43:38.000Z","dependencies_parsed_at":null,"dependency_job_id":"a9a271de-201d-485b-97d7-ab827c719558","html_url":"https://github.com/skent259/ordinal-mil-nnets","commit_stats":null,"previous_names":["skent259/ordinal-mil-nnets"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/skent259%2Fordinal-mil-nnets","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/skent259%2Fordinal-mil-nnets/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/skent259%2Fordinal-mil-nnets/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/skent259%2Fordinal-mil-nnets/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/skent259","download_url":"https://codeload.github.com/skent259/ordinal-mil-nnets/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245632406,"owners_count":20647205,"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":["multiple-instance-learning","neural-network","ordinal","python","reproducible-research","simulation","weakly-supervised-learning"],"created_at":"2025-01-31T11:34:53.774Z","updated_at":"2025-03-26T10:21:46.337Z","avatar_url":"https://github.com/skent259.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Ordinal, Multiple Instance Deep Learning\n\nMany approaches have been proposed for deep learning for ordinal labels or multiple-instance learning (MIL) structure separately. However, few works have proposed a deep learning framework for ordinal labels and MIL structure together. This repository contains tensorflow-based code for implementing a deep learning framework for ordinal, MIL data, and includes experiments from the manuscript \"Ordinal, Multiple Instance Deep Learning\" by Sean Kent and Menggang Yu.\n\nAs a quick reference for getting started:\n\n* The experiments can be run via the `run.sh` file, however this will take considerable CPU time. We recommend that you run in a high-throughput environment in batches (see `run.sh` for inspiration)\n* Raw model code is contained in the `models/` directory, in addition to helper code that allows for running the experiment and application. \n* If you are looking for a quick way to re-use the underlying code for your own deep learning, the examples in the `test/` directory are a useful place to start. \n* Analysis, including code to replicate the figures and tables, is in the `analysis/` directory for the main experiment and the `application/` directory for an application to TMA data. \n* Data (open-source) can be downloaded and processed in individual folders under the `datasets/` directory. \n* The `condor/` directory can be ignored. It was used to run simulations in the HTCondor (high-throughput) environment.\n\n\n## Methods compared\n\nA full description of the methods is given in the manuscript. Where code was not available in a package, but the implementation was present elsewhere, we have copied the code into a directory with credit given below.\n\n| Method name              | Type    | Directory                                                 | Reference     |\n| ------------------------ | ------- | --------------------------------------------------------- | ------------- |\n| mi-net                   | MIL     | `mil_nets/`                                               | [1], [2], [3] |\n| MI-net                   | MIL     | `mil_nets/`                                               | [2]           |\n| MI-net (DS)              | MIL     | `mil_nets/`                                               | [2]           |\n| MI-net (Attention)       | MIL     | `mil_attention/`                                          | [3]           |\n| MI-net (Gated-attention) | MIL     | `mil_attention/`                                          | [3]           |\n| CORAL                    | Ordinal | NA, see \"coral-ordinal\",  \"coral-pytorch\" python packages | [4]           |\n| CORN                     | Ordinal | NA, see \"coral-ordinal\",  \"coral-pytorch\" python packages | [5]           |\n| CLM QWK                  | Ordinal | `clm_qwk/`                                                | [6]           |\n\n* `mil_nets/`: Originally used \u003chttps://github.com/yanyongluan/MINNs\u003e for inspiration/testing, but final code uses layers from tensorflow.\n* `mil_attention/`: \u003chttps://keras.io/examples/vision/attention_mil_classification/\u003e\n* `clm_qwk/` \u003chttps://github.com/ayrna/deep-ordinal-clm\u003e\n* \"coral-ordinal\": \u003chttps://github.com/ck37/coral-ordinal\u003e\n* \"coral-pytorch\": \u003chttps://github.com/Raschka-research-group/coral-pytorch\u003e\n\n## References\n\n[1] Ramon, J., \u0026 De Raedt, L. (2000). Multi instance neural networks. Proceedings of the ICML-2000 Workshop on Attribute-Value and Relational Learning, 53–60.\n\n[2] Wang, X., Yan, Y., Tang, P., Bai, X., \u0026 Liu, W. (2018). Revisiting multiple instance neural networks. Pattern Recognition, 74, 15–24. https://doi.org/10.1016/j.patcog.2017.08.026\n\n[3] Ilse, M., Tomczak, J., \u0026 Welling, M. (2018). Attention-based deep multiple instance learning. Proceedings of the 35th International Conference on Machine Learning, 2127–2136. https://proceedings.mlr.press/v80/ilse18a.html\n\n[4] Cao, W., Mirjalili, V., \u0026 Raschka, S. (2020). Rank consistent ordinal regression for neural networks with application to age estimation. Pattern Recognition Letters, 140, 325–331. https://doi.org/10.1016/j.patrec.2020.11.008\n\n[5] Shi, X., Cao, W., \u0026 Raschka, S. (2022). Deep neural networks for rank-consistent ordinal regression based on conditional probabilities. ArXiv Preprint ArXiv:2111.08851. http://arxiv.org/abs/2111.08851\n\n[6] Vargas, V. M., Gutiérrez, P. A., \u0026 Hervás-Martínez, C. (2020). Cumulative link models for deep ordinal classification. Neurocomputing, 401, 48–58. https://doi.org/10.1016/j.neucom.2020.03.034\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fskent259%2Fordinal-mil-nnets","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fskent259%2Fordinal-mil-nnets","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fskent259%2Fordinal-mil-nnets/lists"}