Added _lupdf docs for normal distribution #277

bbbales2 · 2020-10-21T12:40:38Z

Submission Checklist

Builds locally
Declare copyright holder and open-source license: see below

Summary

Three things:

I moved the probability conventions section out from under the discrete distributions section. It is now below Deprecated functions which seems a bit weird, but it is close to the probability sections without being in one of them.
When I used the section link syntax [linkname](#sectionname) the link bar on the left ended up in the wrong place so I ended up linking to the html file directly via [linkname](sectionname.html) and that seemed to work.
Added a line for _lupdf/_lupmf for all the distributions

Copyright and Licensing

Please list the copyright holder for the work you are submitting (this will be you or your assignee, such as a university or company): Columbia University

By submitting this pull request, the copyright holder is agreeing to license the submitted work under the following licenses:

Code: BSD 3-clause (https://opensource.org/licenses/BSD-3-Clause)
Documentation: CC-BY 4.0 (https://creativecommons.org/licenses/by/4.0/)

mitzimorris · 2020-10-21T18:00:48Z

this is a good way to do it. do you want a script to do this automatically on a file by file basis?

bbbales2 · 2020-10-21T18:53:36Z

Ah this should be easy enough to copy paste. I'll go ahead and finish it up that way.

bbbales2 · 2020-10-21T19:19:01Z

@mitzimorris actually I changed my mind on how to do this. Can you look again? I think this is closer to what was there previously.

mitzimorris · 2020-10-21T21:16:12Z

src/functions-reference/probability_function_conventions.Rmd

@@ -1,4 +1,4 @@
-# Conventions for Probability Functions
+# Probability Function Conventions


please don't change chapter names - it messes up how we do the linking into latest version.
you should also change the filename back to how it was, since filename matches chapter name.

Oh okay, can change this back. Does that mean the new way of doing the _lupdf s is okay?

bbbales2 · 2020-10-21T21:32:10Z

I ran into some dependent vs dependant s.

I did some grepping:

bbales2@tadpole:~/docs$ grep -r "dependent" src/functions-reference/* | wc -l
102
bbales2@tadpole:~/docs$ grep -r "dependant" src/functions-reference/* | wc -l
7

I'm gonna change the dependant s I encounter to dependent.

bbbales2 · 2020-10-21T21:52:07Z

src/functions-reference/binary_distributions.Rmd


 `real` **`bernoulli_logit_glm_lpmf`**`(int y | matrix x, real alpha, vector beta)`<br>\newline
 The log Bernoulli probability mass of y given chance of success
-`inv_logit(alpha + x * beta)`, where the same intercept `alpha` and dependant variable value `y` are used


I cut a lot of the text here about the size constraints. It's unnecessary I think. We aren't explicit about size constraints in other distributions.

bbbales2 · 2020-10-21T23:14:14Z

@rok-cesnovar could you have a look through these? I've done some other cleanup but I think this is vaguely what we want. Easy to go through and change something so don't hesitate to ask.

rok-cesnovar · 2020-10-22T06:33:28Z

The changes look good to me! I think this is enough info. Thanks.

bbbales2 · 2020-10-22T20:11:07Z

Ready for review!

@jgabry here's (edit: an example of) what we did for functions guide:
unbounded-continuous-distributions.html.zip

Somewhat monotonous but not too hard to change things if you think we should add/remove something.

jgabry · 2020-10-22T20:42:13Z

@bbbales2 Thanks. What should I be looking at in unbounded-continuous-distributions.html.zip? When I open it it just looks like the table of contents.

bbbales2 · 2020-10-22T20:44:38Z

Ooops. I was trying to be convenient and failed.
Maybe this: normal-distribution.html.zip

If it doesn't work you'll have to build your own. Just showing how we added _lupdf/_lupmf to the function reference. Basically we did it brute force.

jgabry · 2020-10-22T23:09:41Z

Thanks that one worked, thanks.

I think this looks good for the function documentation. It seems reasonable to just have the _lupdf function right after the _lpdf one and the explanation accompanying it is pretty clear. So this all looks good. I do think we should add more about this stuff in the user guide but we don't have to hold this up for that. I think it's worth going ahead and merging the function reference documentation that you already have here.

SteveBronder · 2020-11-02T17:15:57Z

src/functions-reference/binary_distributions.Rmd

-If `x` and `y` are data (not parameters) this function can be executed on a GPU.
+`inv_logit(alpha + x * beta)`.
+
+If `x` and `y` are data, this function can be executed on a GPU.


@rok-cesnovar we can cut this line out now right?

Indeed we can yes.

Should I remove them for all distributions? (please say yes)

Sir yes sir!

Sir I will remove all of these messages sir!

Sir messages removed sir

Sir edit: sir

SteveBronder

I looked this over and seems good! Only thing I want to check is whether the line

If `x` and `y` are data (not parameters) this function can be executed on a GPU.

for the glm functions on gpus is needed anymore since the new glm's I think can take in parameters for those values

bbbales2 · 2020-11-03T02:07:34Z

@SteveBronder I removed the GPU things, so merge if yah think it looks alright and send @rok-cesnovar a note to do the announcement

rok-cesnovar · 2020-11-03T18:43:55Z

Can we merge. We need to rebuild html before the announcement.

bbbales2 · 2020-11-03T22:57:09Z

@SteveBronder are my latest changes good?

SteveBronder

Good!

Added _lupdf docs for normal distribution

9941bfb

bbbales2 requested a review from mitzimorris October 21, 2020 12:40

Changed how normal_lupdf is presented

07759e8

mitzimorris reviewed Oct 21, 2020

View reviewed changes

bbbales2 added 3 commits October 21, 2020 17:43

Added _lupmfs to binary distributions

b96e0f8

Renamed probability functions conventions chapter

18ff581

Fixed typo

ee0a216

bbbales2 commented Oct 21, 2020

View reviewed changes

Documented bounded distcrete _lupmfs

b803c6c

bbbales2 added 5 commits October 22, 2020 14:18

Part way through updating all densities to have lupmf/lupdf docs

014dff5

Positive continuous distributions _lupdfs

4e3ca2b

Positive lower bounded distribution _lupdfs

3b23b51

Simplex distribution lupdf

79fd0b4

Unbounded continuous distribution lupdfs

0602372

SteveBronder reviewed Nov 2, 2020

View reviewed changes

SteveBronder approved these changes Nov 2, 2020

View reviewed changes

Removed GPU limitations docs on glm functions

e137bbe

SteveBronder approved these changes Nov 4, 2020

View reviewed changes

SteveBronder merged commit aa89620 into master Nov 4, 2020

WardBrian deleted the feature/lupdf-functions-reference branch March 28, 2022 13:46

		@@ -1,4 +1,4 @@
		# Conventions for Probability Functions
		# Probability Function Conventions

Uh oh!

Added _lupdf docs for normal distribution #277

Added _lupdf docs for normal distribution #277

Uh oh!

Conversation

bbbales2 commented Oct 21, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Submission Checklist

Summary

Copyright and Licensing

Uh oh!

mitzimorris commented Oct 21, 2020

Uh oh!

bbbales2 commented Oct 21, 2020

Uh oh!

bbbales2 commented Oct 21, 2020

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bbbales2 commented Oct 21, 2020

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bbbales2 commented Oct 21, 2020

Uh oh!

rok-cesnovar commented Oct 22, 2020

Uh oh!

bbbales2 commented Oct 22, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jgabry commented Oct 22, 2020

Uh oh!

bbbales2 commented Oct 22, 2020

Uh oh!

jgabry commented Oct 22, 2020

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bbbales2 Nov 2, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

SteveBronder left a comment

Choose a reason for hiding this comment

Uh oh!

bbbales2 commented Nov 3, 2020

Uh oh!

rok-cesnovar commented Nov 3, 2020

Uh oh!

bbbales2 commented Nov 3, 2020

Uh oh!

SteveBronder left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

bbbales2 commented Oct 21, 2020 •

edited

Loading

bbbales2 commented Oct 22, 2020 •

edited

Loading

bbbales2 Nov 2, 2020 •

edited

Loading