Basketball-rctivity-recognition

Project for recognizing the different activities of two subjects.

1. Data collection using the 3 wearable sensors.
2. Data preprocessing and applying sliding window for time dependent activites.
3. Applying LSTM neural network with 'n' k-fold for better model validation.
4. Applying KNN classifier with 'n' k-fold for better model validation.

Coding

1. The file named "HAR_LSTM" is the code file starts from reading all the .csv files.
2. I had a video with time running on it to find the matching data from start to end time of each activity. (Since the video size is bit larger, I could not provide in my repository. But I have mentioned the exact starting time of each activity in the code with frquency which will end exactly with matching time between video playback and data from .csv )
3. Feel free to contact me any time.....Thank you