So there are a couple of ways to go about this.
In a Python shell, you can use the package 'pandas'. I personally enjoy the ease of using Jupyter Notebook. You can access this by installing anaconda PowerShell and then within the PowerShell type "Jupiter notebook" and it will launch within your working directory. Or, you can use a different ...
From your question, you are using a swath dataset (i.e. not using a regular grid). If you were using a regular grid, you'd just calculate the pixel number from the e.g. corner coordinates. I think
geolocation = Dataset(fname)
points = np.array([geolocation.variables['lon'][:],
# Exact ...