https://github.com/nwhovian/activity-recognition
Human Activity Recognition using mobile sensors
https://github.com/nwhovian/activity-recognition
human-activity-recognition python
Last synced: 3 months ago
JSON representation
Human Activity Recognition using mobile sensors
- Host: GitHub
- URL: https://github.com/nwhovian/activity-recognition
- Owner: nWhovian
- License: mit
- Created: 2020-04-21T22:06:04.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2020-04-21T22:13:49.000Z (about 5 years ago)
- Last Synced: 2025-01-10T21:24:24.860Z (5 months ago)
- Topics: human-activity-recognition, python
- Language: Python
- Size: 4.88 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Human Activity Recognition using mobile sensors
In my algorithm, I used the previously pre-processed and divided into train/test sections data set and created three models using the scikit-learn library. Then I have tuned the hyper-parameters of these models using the Grid-Search-CV and came up with the best set of parameters for each model. Finally, I have evaluated each model on the test data set and created a confusion matrix for it.
## Results
```
sgd average accuracy: 95.374
log_reg average accuracy: 96.132
svm average accuracy: 95.792
```
```
sgd average exec time: 70.502 s
log_reg average exec time: 144.65 s
svm average exec time: 75.824 s
```