Machine learning models that predict human activity upon receiving sensors readings
-
Updated
Aug 27, 2022 - Jupyter Notebook
#
pamap2
Here are 4 public repositories matching this topic...
Classifies football players into Attacker, Midfielder, Defender roles from PAMAP2 IoT wearable data. LSTM, BiLSTM, and TCN-Transformer architectures. 99.24% accuracy, LOSO 98.89%±0.42%. SHAP sensor attribution per role.
iot activity-recognition pytorch transformer lstm data-analytics football tcn bilstm sports-analytics shap positional-analysis football-analytics wearable-sensors pamap2 loso player-role-classification
-
Updated
Apr 21, 2026 - Jupyter Notebook
Three-class football player fatigue prediction from PAMAP2 wearable IoT data. Karvonen heart rate labeling, SMOTE balancing, LOSO cross-validation, personalized Random Forest. 97.96% LOSO accuracy + coach substitution-alert dashboard.
iot data-science random-forest tensorflow scikit-learn lstm data-analytics football smote sports-analytics football-analytics wearable-sensors pamap2 injury-prevention loso coach-dashboard fatigue-prediction karvonen
-
Updated
Apr 21, 2026 - Jupyter Notebook
Human Activity Recognition using wearable sensor data (PAMAP2 dataset). Implements classical machine learning algorithms from scratch including Logistic Regression, KNN, SVM, Least Squares, PCA, and more.
-
Updated
May 5, 2026 - Jupyter Notebook
Improve this page
Add a description, image, and links to the pamap2 topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the pamap2 topic, visit your repo's landing page and select "manage topics."