A cool way of visualising likelihood peaks of the parameters of a Gaussian GLM as a 3D mesh. You can visualise it as the "normal" standard likelihood, or as the loglikelihood.
Just insert two variables into the model.
You may see how the coordinates of the peaks correspond to the approximations of intercept and slope made by sklearn.
In the file uploaded I used my own files. Please feel free to use any two variables.
Made on Python 3.8.16, Jupyter Notebook.