MotionSense Dataset for Human Activity and Attribute Recognition ( time-series data generated by smartphone's sensors: accelerometer and gyroscope) (PMC Journal) (IoTDI'19)
Thanks for the dataset! In the description of the data set, it is said that A_DeviceMotion_data contains B_Accelerometer_data and C_Gyroscope_data, but I found that the number of files corresponding to each of the three data sets is not exactly the same, and the data values are not the same. Could you please tell me how to get A from BC?
I found that g was asked in the question area. If I use A for HAR in deep model, can I use A instead of B and C? At the same time, do I need to combine g to transform the data? But I feel that the constant multiplied here is meaningless after normalization in the later stage.
The data is segmented in the collection, are they continuous over the course of acquisition? For example, whether the data of sub_1 in dws1 is adjacent to dws2 and ups_3...... during the collection process sub_1 in?
Thanks for the dataset! In the description of the data set, it is said that A_DeviceMotion_data contains B_Accelerometer_data and C_Gyroscope_data, but I found that the number of files corresponding to each of the three data sets is not exactly the same, and the data values are not the same. Could you please tell me how to get A from BC?
I found that g was asked in the question area. If I use A for HAR in deep model, can I use A instead of B and C? At the same time, do I need to combine g to transform the data? But I feel that the constant multiplied here is meaningless after normalization in the later stage.
The data is segmented in the collection, are they continuous over the course of acquisition? For example, whether the data of sub_1 in dws1 is adjacent to dws2 and ups_3...... during the collection process sub_1 in?
Thanks a lot！