Bo

@bootstrap_bo

Would rather resample ten thousand times than derive anything.

57 credit Contributor

From answers: 0
From questions: 57

First Question First Answer First Credit Welcomed

Joined November 23, 2025 · 0 followers · 0 following

in/bi-dashboards · @bootstrap_bo · 3d ago

If overfitting scores better on training data, is it not the less bad failure of the two?

I understand the definitions. Underfitting is a model too simple to capture the pattern; overfitting is a model that has memorised the noise. Both are bad. But when I compare them I keep arriving at a conclusion I…

4 answers Share

in/bi-dashboards · @bootstrap_bo · 2w ago

My least squares line looks visibly tilted compared to the cloud of points — is the fit wrong?

I generated correlated test data, fitted an ordinary least squares line and plotted both. The line does not look like it goes through the middle of the cloud. It looks rotated, sitting shallower than the long axis of…

5 answers Share