{"id":22968420,"url":"https://github.com/irc-sphere/embedded-features","last_synced_at":"2025-06-25T06:04:50.856Z","repository":{"id":71292358,"uuid":"114171747","full_name":"IRC-SPHERE/embedded-features","owner":"IRC-SPHERE","description":"Source code release for our EWSN 2018 paper","archived":false,"fork":false,"pushed_at":"2018-10-17T11:47:24.000Z","size":5429,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":7,"default_branch":"master","last_synced_at":"2025-04-02T05:25:41.599Z","etag":null,"topics":["acceleration-data","machine-learning"],"latest_commit_sha":null,"homepage":null,"language":"C","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/IRC-SPHERE.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":"2017-12-13T21:40:58.000Z","updated_at":"2019-08-28T22:03:06.000Z","dependencies_parsed_at":"2024-02-21T07:00:31.149Z","dependency_job_id":null,"html_url":"https://github.com/IRC-SPHERE/embedded-features","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/IRC-SPHERE/embedded-features","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IRC-SPHERE%2Fembedded-features","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IRC-SPHERE%2Fembedded-features/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IRC-SPHERE%2Fembedded-features/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IRC-SPHERE%2Fembedded-features/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/IRC-SPHERE","download_url":"https://codeload.github.com/IRC-SPHERE/embedded-features/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/IRC-SPHERE%2Fembedded-features/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":261816312,"owners_count":23213863,"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":["acceleration-data","machine-learning"],"created_at":"2024-12-14T21:19:09.089Z","updated_at":"2025-06-25T06:04:50.845Z","avatar_url":"https://github.com/IRC-SPHERE.png","language":"C","funding_links":[],"categories":[],"sub_categories":[],"readme":"This is a test of feature extraction performance (speed, RAM usage, ROM usage) on ARM Cortex-M3 and M4 cores.\n\nPlease cite the following paper if you use this repository:\n\n* A. Elsts, R. McConville, X. Fafoutis, N. Twomey, R. Piechocki, R. Santos-Rodriguez and I. Craddock. On-Board Feature Extraction from Acceleration Data for Activity Recognition, EWSN 2018.\n\n### Features\n\nImporting data:\n\n    cd import\n    ./load_sphere_challenge_files.py\n\nThis creates a number of `.c` files in the `data` directory, each containing 15000 samples of 3-axis acceleration data. The data is expressed as 8-bit signed integers.\n\n### Running\n\nRunning natively (the same architecture as on the machine it is compiled on):\n\n    cd feature-test\n    make run ARCHITECTURE=native\n\n\nRunning on an emulator - needs Zephyr OS and Zephyr SDK to be installed:\n\n    cd feature-test\n    make run ARCHITECTURE=zephyr\n\n\nRunning on hardware (a SPHERE board) - needs Contiki OS and ARM compiler to be installed:\n\n    cd feature-test\n    make run ARCHITECTURE=sphere\n    \nOther supported targets are: `z1` (for Zolertia Z1 with msp430), `zoul` (for Zolertia Zoul with CC2538) and `nrf52dk` (for Nordic NRF52DK with Cortex-M4F).\n\n\n### Features\n\nNote: for increased performance, some of the FFT-based feature results are not normalized!\nTo get the correct result, they should be divided either by `FREQUENCY_FEATURE_WINDOW_SIZE` or by `FREQUENCY_FEATURE_WINDOW_SIZE^2` depending on the feature.\n\n\n### Generating results for comparison\n\nGo to the top directory and run\n\n    ./generate_features.py\n\nThis will generate a number of CSV files in the `export` directory.\nEach file is based on 25-minute data from a single participant and includes all of the supported features.\n\nThis will run the feature test natively. Prerequisites:\n\n* GNU make\n* GCC\n* Python3\n\n### Results\n\n* The `export` directory will contain the `.csv` files with the different features after running `generate_features.py`.\n* The `result` directory contains the processing duration evaluation results on the different hardware platforms.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Firc-sphere%2Fembedded-features","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Firc-sphere%2Fembedded-features","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Firc-sphere%2Fembedded-features/lists"}