{"id":13571425,"url":"https://github.com/kaist-dmlab/TCLP","last_synced_at":"2025-04-04T08:31:08.408Z","repository":{"id":39442409,"uuid":"468699062","full_name":"kaist-dmlab/TCLP","owner":"kaist-dmlab","description":"Implementation of ICLR'22 Coherence-based Label Propagation over Time Series for Accelerated Active Learning","archived":false,"fork":false,"pushed_at":"2022-12-19T06:04:54.000Z","size":124,"stargazers_count":19,"open_issues_count":1,"forks_count":17,"subscribers_count":3,"default_branch":"main","last_synced_at":"2024-11-05T04:34:06.366Z","etag":null,"topics":[],"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/kaist-dmlab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2022-03-11T10:07:22.000Z","updated_at":"2024-02-02T03:24:04.000Z","dependencies_parsed_at":"2023-01-29T20:46:00.955Z","dependency_job_id":null,"html_url":"https://github.com/kaist-dmlab/TCLP","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/kaist-dmlab%2FTCLP","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kaist-dmlab%2FTCLP/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kaist-dmlab%2FTCLP/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kaist-dmlab%2FTCLP/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/kaist-dmlab","download_url":"https://codeload.github.com/kaist-dmlab/TCLP/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247147085,"owners_count":20891619,"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":"2024-08-01T14:01:01.911Z","updated_at":"2025-04-04T08:31:07.744Z","avatar_url":"https://github.com/kaist-dmlab.png","language":"Python","readme":"\nThis github is the implementation of ICLR'22 paper, named as **Coherence-based Label Propagation over Time Series for Accelerated Active Learning**.\nPlease follow the instructions to reproduce our work.\n\n\n# Download python packages in requirements.txt\n```shell \npip install -r requirements.txt\n```\n\n# Download datasets\nDownload I3D feature data of 50salads and GTEA at [link](https://zenodo.org/record/3625992#.YVwLbdpBx1N) and locate the contents at the `./datasets/DATASET_NAME`. The link is from [ms-tcn](https://github.com/yabufarha/ms-tcn), the repository for the paper \"Y. Abu Farha and J. Gall. MS-TCN: Multi-Stage Temporal Convolutional Network for Action Segmentation. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.\" \nFor HAPT and mHealth dataset, download [HAPT](http://archive.ics.uci.edu/ml/datasets/smartphone-based+recognition+of+human+activities+and+postural+transitions) and [mHealth](http://archive.ics.uci.edu/ml/datasets/mhealth+dataset) and locate the contents at the `./datasets/DATASET_NAME`\n\n# How to run\n\nAt current directory which has all source codes, run main.py with the parameters as follows. \n\n\t- dataset: {50salads, GTEA, mHealth, HAPT} designates which dataset to use.\n\t- seed: {0, 1, 2, 3, 4} is the seed for 5-fold cross validation.\n\t- gpu: {0, 1, 2, ...} is an integer for gpu id\n\t- lp: {platprob, repr, prob, zero} indicates label propagation method to use, representing {TCLP, ESP, PTP, NOP} in the paper, respectively.\n\t- al: {conf, entropy, margin, core, badge, utility}\tshows active learning to use, representing {CONF, ENTROPY, MARG, CS, BADGE, UTILITY} in the paper, respectively.\n\t- no_plat_reg: {0, 1}\tdecides whether or not to use width regularization or not. 1 means removing width regularization.\n\t- temp: [1, infinity] is the parameter T for temperature scaling. T=1 means no temperature scaling.\n\nHere's the example running code.\n\n```shell\npython3 main.py --dataset HAPT --gpu 0 --seed 0 --lp platprob --al random --no_plat_reg 1 --temp 2.0\n```\nClassification accuracy at each active learning round is saved in `metadata` folder as `.npy` format.\n\n# Citation\n\nPlease use the following form to cite our paper.\n\n```\n@inproceedings{\nshin2022coherencebased,\ntitle={Coherence-based Label Propagation over Time Series for Accelerated Active Learning},\nauthor={Yooju Shin and Susik Yoon and Sundong Kim and Hwanjun Song and Jae-Gil Lee and Byung Suk Lee},\nbooktitle={International Conference on Learning Representations},\nyear={2022},\nurl={https://openreview.net/forum?id=gjNcH0hj0LM}\n}\n```\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n","funding_links":[],"categories":["📝 Papers with code"],"sub_categories":["Managed database services"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkaist-dmlab%2FTCLP","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkaist-dmlab%2FTCLP","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkaist-dmlab%2FTCLP/lists"}