-
-
Notifications
You must be signed in to change notification settings - Fork 2.1k
The Gumbel Distribution is incorrectly described #6478
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.
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
The Gumbel Distribution is incorrectly described #6478
Conversation
The version currently in PYTensor is the Gumbel Right Skew which gives the extreme maximum. Anyone using this to find the extreme minimum--in survival analysis for example, is using the wrong distribution. A sign change on the x and y inputs is an effective solution if the user is notified before hand.
Univariate Gumbel right Skew log-likelihood. This parameterization will provide the extrem maximum value. | ||
Those looking to find the extrem minimum provided by the left skew Gumbel should | ||
invert the sign to all X and Y values (positive to negative and negative to positive). |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Univariate Gumbel right Skew log-likelihood. This parameterization will provide the extrem maximum value. | |
Those looking to find the extrem minimum provided by the left skew Gumbel should | |
invert the sign to all X and Y values (positive to negative and negative to positive). | |
Univariate Gumbel right Skew log-likelihood. This parameterization will provide the extreme maximum value. | |
Those looking to find the extreme minimum provided by the left skewed Gumbel should | |
invert the sign to all X and Y values (positive to negative and negative to positive). |
fixes two typos and removes trailing whitespace which our pre-commit
would complain about
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It also needs to be formatted as 1st sentence, a white line and then the more extense description to follow numpydoc: https://numpydoc.readthedocs.io/en/latest/format.html#sections
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hell, sorry for not looking at these comments. I am dealing with wedding planning and work at the same time. will look into this around the end of February.
Anyone feels like picking this one up? |
It looks like it is only missing the corrections pointed out above. Let me know if I missed anything. |
Closing in favor of #6810. |
The version currently in PYTensor is the Gumbel Right Skew which gives the extreme maximum. Anyone using this to find the extreme minimum--in survival analysis for example, is using the wrong distribution. A sign change on the x and y inputs is an effective solution if the user is notified before hand.
What is this PR about?
...
Checklist
Major / Breaking Changes
New features
Bugfixes
Documentation
Univariate Gumbel right Skew log-likelihood. This parameterization will provide the extrem maximum value.
Those looking to find the extrem minimum provided by the left skew Gumbel should
invert the sign to all X and Y values (positive to negative and negative to positive).
Maintenance