Report
1 year, 3 months ago.

How to make K-Neareset Neighbor possible for nRF51822_ADXL345 using arduino MEGA code

Hello World... I wanna share with my new progress to implement machine learning for nRF51822. Here it is how to make it possible. after questioning to forum that using machine learning is hardfull to implemented to it. So here's some advice from devzone forum.

You have now 2 options: 1. Transfer the data to something bigger like a PC or mobile and do the AI there. 2. Ditch the AI and use a much less expensive classic statistics algorithm.

I wanna to make it possible so, after talking about how to implement this. first we have to know how to use it. first of all we know machine learning using 3 method that was preparation data, training and recognition. So because Preparation and training mode that're using large RAM and calculation so I have to use python for the best solution KNN python1, KNN python 2. So here's the image..

/media/uploads/asyrofi/python_knn1.png /media/uploads/asyrofi/python_knn2.png

that's I'm using it from several method to use knn using python.. after that I wanna use that KNN pattern to implement for recognition. So I have using from that KNN arduino code. So this is the result.

/media/uploads/asyrofi/arduino_knn1.png /media/uploads/asyrofi/arduino_knn2.png

ok, that's is the result how I using it, so if I'm using mbed online compiler and how to make it possible. please help me to do it. I'm so appreciate w/ your help please.. and another thing if using this one utensor. How to make it possible for nRF51822 which is have 16kb, but that library minimum using 256kb. please help me soon..

this is my mbed code..

Comment on this question
Be the first to answer this question.

Assigned to rakha asyrofi 1 year, 3 months ago.

This means that the question has been accepted and is being worked on.

You need to log in to post a question