meanshift | meanshift object tracking algorithm in C

Hello I am a machine learning newbie. I need some help with unsupervised clustering of high dimentional data. I have data with over 15 dimensions with around 50 - 80 thousand rows. The data looks something like this (15 participants with almost equal number of rows each and 15 features) -

The data consists of many participants, each participant has multiple rows of data and they are time stamped with their features. My goal is to cluster this data according to participants and make inferences based on these clusters. The problem here is that there are many rows for each participant and I cannot represent each participant with a single point so clustering them seems like a difficult task.

I would like help with:

1. What would be the best way to cluster this data so that I can make inferences according to the participant ?

2. Which clustering technique should I use? I have tried sklearn's Kmeans, meanshift and other libraries but they take too long and crash my system.

Sorry If it's a bit difficult to understand I will try my best to answer your questions. Thank you in advance for the help. If this question is very similar to some other question please let me know (I was not able to find it).

Thank you :)

I am writing an inference system for a underwater vehicle and I've been told to look into async to improve the object tracking. The idea is to have a tflite object detection model that detects object and returns a box/coordinates for each object it detects, which then meanshift (or some other tracking algo then uses to track the objects). However, the tflite model takes roughly 1 second to do its detection which is too slow. So I want to be running that in a separate thread while meanshift is tracking, whenever the tflite model is done, it updates the box to track. That way I would assume we would have smooth detection and tracking.

I find async kinda tricky and cannot quite get it right. For testing purposes I created a inference function with a 5 second delay just to clearly emulate the time inference takes and a tracking function that just runs continuously to emulate meanshift. This is what I have so far:

I'm trying to install scikit-learn with pip by using pip install scikit-learn and I got this message:

DEPRECATION: Python 2.7 reached the end of its life on January 1st, 2020. Please upgrade your Python as Python 2.7 is no longer maintained. pip 21.0 will drop support for Python 2.7 in January 2021. More details about Python 2 support in pip can be found at https://pip.pypa.io/en/latest/development/release-process/#python-2-support pip 21.0 will remove support for this functionality.

WARNING: The scripts f2py, f2py2 and f2py2.7 are installed in '/Users/my_name/Library/Python/2.7/bin' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location.

However, when I type python --version it says that my version is 3.7.4. Are python 2.7 and 3.7 both installed on my computer then? If so, is there a way I can get rid of 2.7? Also, I still get a ModuleNotFoundError when I do from sklearn.cluster import MeanShift in the mu_editor. I think it's because pip is installing the modules to the python 2.7 directory or something, instead of the python 3.7.4 directory that the mu_editor is connected to. Is there any way I can install the packages to the python 3.7.4 directory instead of the 2.7 one?

I am not sure if it is the bandwith estimator or Meanshift that is taking ages, I want to know how i can reduce the time, maybe I need to modify the image or maybe the parameters of the the clustering algo.

THIS IS THE IMAGE:

THIS IS MY CODE:

I have a working example of Mean Shift clustering using Pandas and Sci-kit learn. I am new to Python so I think I am missing something basic here. Here is my working code:

I am using built-in dataset iris from sklearn for clustering. In KMeans I set the number of clusters in advance but it is not true for DBSCAN. How to train a model if you dont set the number of clusters in advance?

I tried:

I'm trying to get the frame first, so I can identify a face from it, (using my computer webcam) but when I'm calling the read() method, it crashes. saying tuple index out of range. I figured that's because the first frame I give it to detect from is empty. so it has nothing to turn into a tuple. but the guy im learning from (course my teacher gave me), he does it, and it works for him.

my code: