New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Sweep score calculation for tp and fp #366

Open

szympie opened this issue May 26, 2020 · 1 comment

szympie commented May 26, 2020

During calculation of sweep score, if the anomalyPoint is in the anomaly windows you divide by windowWidth (https://github.com/numenta/NAB/blob/master/nab/sweeper.py#L176) and if the anomalyPoint is outside the anomaly window, you divide by windowWidth - 1 (https://github.com/numenta/NAB/blob/master/nab/sweeper.py#L188) is there any reason for it or is it a bug?

Contributor

markNZed commented Oct 22, 2021

I agree that it seems strange to change the window width. In the window the position formula is

-(curWindowRightIndex - i + 1) / curWindowWidth

After the window the formula is

abs(prevWindowRightIndex - i) / float(prevWindowWidth - 1)

I doubt it makes an important difference to the scores.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment