NeuralNet-HumanActivityRecognition
This program uses 3 layer fully connected neural network to detect human activity with approximately 98% train and 96% test accuracy.
Dependencies:
- Numpy
All the details about parameters and Algorithm are clearly mentioned along with the code. The train and test datasets can be found in # Activity folder.
More details about the problem statement can be found here https://www.kaggle.com/uciml/human-activity-recognition-with-smartphones