Activity Recognition from Single Chest-Mounted Accelerometer:
Uncalibrated Accelerometer Data are collected from 15 participants performing 7 activities. The dataset provides challenges for identification and authentication of people using motion patterns.
Diffusion accelerometerEBClassifier (master)
Recent Commits
Recent Commits
Commit | Author | Details | Committed | ||||
---|---|---|---|---|---|---|---|
de3567508a43 | zanoli | readme mod and clean jupyter notebook | Feb 18 2020 | ||||
cbd6d213df84 | zanoli | readme, mod | Feb 18 2020 | ||||
ca565edca3c9 | zanoli | readme | Feb 18 2020 | ||||
c06d05f5af95 | zanoli | testing library | Feb 17 2020 | ||||
af515fdfeb26 | zanoli | First results | Feb 14 2020 | ||||
1fd0ce906566 | zanoli | starting using the library | Feb 13 2020 | ||||
7e3e48d0bbcd | zanoli | minor typos | Feb 11 2020 | ||||
6fdf3a5ea138 | zanoli | testing the inertial module in the event base lib | Feb 10 2020 | ||||
0eeb0c108d03 | zanoli | minor tests on the DFT | Feb 6 2020 | ||||
c4ece33fd3ea | zanoli | continue | Jan 30 2020 | ||||
259d14d0efa0 | zanoli | gitIgnore having trouble working | Jan 30 2020 | ||||
1866d2b9daf7 | zanoli | asdsagfdh | Jan 30 2020 | ||||
c4979d0b7806 | zanoli | updated gitignore | Jan 30 2020 | ||||
6031cde82ac5 | zanoli | parsed file | Jan 30 2020 |
readme.md
readme.md
This is the main repository for the inertial project. The data for this repository can be fond here: Dataset In order to be able to use this repository:
- create a root directory for this project
- create a root directory for the eventBasedLibrary
- move to the eventBasedLibrary library directory
- clone the eventBasedLibrary repository: git clone ssh://git@c4science.ch/source/eb_lib.git
- run the setup.py script in the eventBased library: "python3 setupy.py"
- install the library: pip install -m "local/path/to/eventBasedLibrary"
- move to this project directory and set it as home directory
- clone this project: git clone ssh://git@c4science.ch/source/acebc.git
- create a data directory: "./data"
- download the dataset, extract it and save it in the data directory
- move to "./preProc"
- run: "python3 reWriteData.py" this script will create a "./data/dataLabeld/" directory with the data saved in a convinient form for the next step
- run "python3 extractAndSub.py" this will create a "./data/dataEB/" that will hold all the directories for the event based sampled signal, divided into x,y and z component. For example: we can have "./data/dataEB/sub4/7gyroz" where "sub4" means that this folder hold the dataset re-sampled with "extractAndSub.py" with a threshold of 4. 7gyroz means that this is the 7th experiment, the sensor is the gyroscope and the axis is the z axis
- Start to execute a jupyter notebook
- open "Untitled.ipynb"
- we can modify the number of analyzed files by changing the variable "numAnalyzeFile = -1" in the second cell
- we can modify the selected level of subsampling changing the variable "sub = 72" in the second cell
- we can modify the ammount of file used for train by changing the variable "percTrainTest = 0.8" in the second cell
- run the notebook
c4science · Help