Re: [R] Integer division

[Göran Broström]

Thanks Richard,

the "rounding claim" was my mistake (as I replied to Martin), I should 
said "truncates toward zero" as you explain.


However, my point was that these two mathematical functions should be 
defined in the documentation, as you also say. And I was surprised that 
there is no consensus regarding the definition of such elementary functions.


Göran

On 2022-12-20 03:01, Richard O'Keefe wrote:

The Fortran '08 standard says <<
One operand of type integer may be divided by another operand of type 
integer. Although the mathematical
quotient of two integers is not necessarily an integer, Table 7.2 
specifies that an expression involving the division
operator with two operands of type integer is interpreted as an 
expression of type integer. The result of such an
operation is the integer closest to the mathematical quotient and 
between zero and the mathematical quotient

inclusively. >>
Another way to say this is that integer division in
Fortran TRUNCATES towards zero.  It does not round and
never has.

C carefully left the behaviour of integer division (/)
unspecified, but introduced the div(,) function with the
same effect as Fortran (/).  Later versions of the C
standard tightened this up, and the C17 standard reads <<
The result of the / operator is the quotient from the division of the 
first operand by the second; the
result of the % operator is the remainder. In both operations, if the 
value of the second operand is

zero, the behavior is undefined.
When integers are divided, the result of the / operator is the algebraic 
quotient with any fractional
part discarded. 107) If the quotient a/b is representable, the 
expression (a/b)*b + a%b shall equal a ;

otherwise, the behavior of both a/b and a%b is undefined.>>

That is, C17 TRUNCATES the result of division towards
zero.  I don't know of any C compiler that rounds,
certainly gcc does not.


The Java 15 Language Specification says
<< Integer division rounds toward 0. >>
which also specified truncating division.


The help for ?"%/%" does not say what the result is.
Or if it does, I can't find it.  Either way, this is
a defect in the documentation.  It needs to be spelled
out very clearly.
R version 4.2.2 Patched (2022-11-10 r83330) -- "Innocent and Trusting"
 > c(-8,8) %/% 3
[1] -3  2
so we deduce that R *neither* rounds *not* truncates,
but returns the floor of the quotient.
It is widely argued that flooring division is more
generally useful than rounding or truncating division,
but it is admittedly surprising.

On Tue, 20 Dec 2022 at 02:51, Göran Broström > wrote:


I have a long vector x with five-digit codes where the first digit of
each is of special interest, so I extracted them through

  > y <- x %/% 1

but to my surprise y contained the value -1 in some places. It turned
out that x contains -1 as a symbol for 'missing value' so in effect I
found that

  > -1 %/% 1 == -1

Had to check the help page for "%/%", and the first relevant comment I
found was:

"Users are sometimes surprised by the value returned".

No surprise there. Further down:

‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’ indicates
       integer division.  It is guaranteed that

       ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding error)

I did expect  (a %/% b) to return round(a / b), like gfortran and gcc,
but instead I get floor(a / b) in R. What is the reason for these
different definitions? And shouldn't R's definition be documented?

Thanks, Göran

__
R-help@r-project.org  mailing list --
To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help

PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html

and provide commented, minimal, self-contained, reproducible code.



__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Amazing AI

[avi.e.gross]

Boris,

What you are telling us is not particularly new or spectacular in a sense.
It has often been hard to grade assignments students do when they choose an
unexpected path. I had one instructor who always graded my exams (in the
multiple courses I took with him) because unlike most of the sheep, I did
not bother memorizing the way something was done or proven and created my
own solutions on the fly during the test, often in ways he had to work hard
at following and that almost amused him as mine tended to be correct, albeit
not what he would have thought of.

Your issue is not particularly about R as similar scenarios can be found in
many languages and environments.

In programming, it is arguably worse as it is common to be able to do things
so many ways. But plain old R has so many packages available, often with
source code, that any student who finds one that does something they want,
may be able to copy and modify some of the functions involved into their own
code already and fool an evaluator into thinking they did it on their own.
That is a tad harder, as many packages improve the code efficiency by
(re)writing many parts in C/C++.

I have seen things like a GUI that lets you click on various check boxes and
other controls and then use those instructions to read in data files, do
various operations on them, and provide output. Some allow quite a bit of
functionality and also offer you the opportunity to see the R code it
generates, and let you adjust that if it does not quite meet your needs.
Much of it involves including various packages and calling functions in
them, but if your students are allowed to use such things, how would you
know how little actual work they did?

I echo what someone else wrote. Training students for their future jobs in
an uncertain and changing future, may be more effective in teaching them how
to find ever better or different ways to get things done, or even switch to
growing areas near their field.  All kinds of automation of jobs are
happening and will continue to happen in the knowledge professions. People
who read manuals cover to cover and keep consulting them constantly are a
rarity. Many people often first do web searches or consult experts including
online versions of the documentation. Many will happily use software that
lets them do more and more with fewer lines of code written by them and
especially when that software has been used and tested long enough to be
relatively free of bugs when used as directed. Why would anyone these days
want to constantly re-invent the wheel and write routines to read in data
from files using known formats when you can use ones that exist and, if
needed in special cases, make some tweaks such as converting a column it
made into integer, back into the floating point you want for some later
reason?

But if your students are using something that is error-prone when used the
way they are using it, that is a problem as they are not only not learning
some basics or techniques you want them to know, but relying on what may be
bad tools without taking the time and effort to check the result or make
their own tweaks. Such software may not provide a way to do something like
treat multiple entries of various kinds as being NA, as an example. So you
would need your own code to check the result between some steps and do your
own further conversions so that "." and "" and "NA" and "-" all become NA,
again, just a made up example.

Yes, some students will easily fool you when grading but that already
happens when someone hires out getting some work done and claims it as their
own.

-Original Message-
From: R-help  On Behalf Of Boris Steipe
Sent: Monday, December 19, 2022 3:16 PM
To: Milan Glacier 
Cc: r-help@r-project.org
Subject: Re: [R] Amazing AI

Exactly. But not just "error prone", rather: eloquently and confidently
incorrect. And that in itself is a problem. When I evaluate students' work,
I implicitly do so from a mental model of the student - aptitude, ability,
experience, language skills etc. That's useful for summative assessment,
since it helps efficiency - but that won't work anymore. I see a need to
assess much more carefully, require fine-grained referencing, check every
single fact ... and that won't scale. And then there is also the spectre of
having to decide when this crosses the line to "concoction" - i.e. an actual
academic offence ...

Best,
Boris



> On 2022-12-19, at 03:58, Milan Glacier  wrote:
> 
> [You don't often get email from n...@milanglacier.com. Learn why this 
> is important at https://aka.ms/LearnAboutSenderIdentification ]
> 
> On 12/18/22 19:01, Boris Steipe wrote:
>> Technically not a help question. But crucial to be aware of, especially
for those of us in academia, or otherwise teaching R. I am not aware of a
suitable alternate forum. If this does not interest you, please simply
ignore - I already know that this may be somewhat OT.
>> 
>> Thanks.
>> 

Re: [R] Integer division

[Richard O'Keefe]

The Fortran '08 standard says <<
One operand of type integer may be divided by another operand of type
integer. Although the mathematical
quotient of two integers is not necessarily an integer, Table 7.2 specifies
that an expression involving the division
operator with two operands of type integer is interpreted as an expression
of type integer. The result of such an
operation is the integer closest to the mathematical quotient and between
zero and the mathematical quotient
inclusively. >>
Another way to say this is that integer division in
Fortran TRUNCATES towards zero.  It does not round and
never has.

C carefully left the behaviour of integer division (/)
unspecified, but introduced the div(,) function with the
same effect as Fortran (/).  Later versions of the C
standard tightened this up, and the C17 standard reads <<
The result of the / operator is the quotient from the division of the first
operand by the second; the
result of the % operator is the remainder. In both operations, if the value
of the second operand is
zero, the behavior is undefined.
When integers are divided, the result of the / operator is the algebraic
quotient with any fractional
part discarded. 107) If the quotient a/b is representable, the expression
(a/b)*b + a%b shall equal a ;
otherwise, the behavior of both a/b and a%b is undefined.>>

That is, C17 TRUNCATES the result of division towards
zero.  I don't know of any C compiler that rounds,
certainly gcc does not.


The Java 15 Language Specification says
<< Integer division rounds toward 0. >>
which also specified truncating division.


The help for ?"%/%" does not say what the result is.
Or if it does, I can't find it.  Either way, this is
a defect in the documentation.  It needs to be spelled
out very clearly.
R version 4.2.2 Patched (2022-11-10 r83330) -- "Innocent and Trusting"
> c(-8,8) %/% 3
[1] -3  2
so we deduce that R *neither* rounds *not* truncates,
but returns the floor of the quotient.
It is widely argued that flooring division is more
generally useful than rounding or truncating division,
but it is admittedly surprising.

On Tue, 20 Dec 2022 at 02:51, Göran Broström  wrote:

> I have a long vector x with five-digit codes where the first digit of
> each is of special interest, so I extracted them through
>
>  > y <- x %/% 1
>
> but to my surprise y contained the value -1 in some places. It turned
> out that x contains -1 as a symbol for 'missing value' so in effect I
> found that
>
>  > -1 %/% 1 == -1
>
> Had to check the help page for "%/%", and the first relevant comment I
> found was:
>
> "Users are sometimes surprised by the value returned".
>
> No surprise there. Further down:
>
> ‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’ indicates
>   integer division.  It is guaranteed that
>
>   ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding error)
>
> I did expect  (a %/% b) to return round(a / b), like gfortran and gcc,
> but instead I get floor(a / b) in R. What is the reason for these
> different definitions? And shouldn't R's definition be documented?
>
> Thanks, Göran
>
> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Amazing AI

[Christopher W. Ryan via R-help]

In clinical medicine, the question the patient asks rarely represents
their main concern. Most of what I've done in my career, and most of
what I've taught, is about how to have the back-and-forth dynamic dialoq
with the patient, to help them formulate what's really on their mind,
and make sure I understand it before proceeding.

Seems to me that statistical consulting, or working IT in an
organization where one is serving "internal" customers, is similar.
Students need to learn the skills of clarifying, verifying, and
paraphrasing a client's needs.

Sure, ChatGPT may be able to generate decent code to do what a client
*says* they want to do, but is that *really* what they are looking for?

So don't retire yet :)

--Chris Ryan

Boris Steipe wrote:
> Technically not a help question. But crucial to be aware of, especially for 
> those of us in academia, or otherwise teaching R. I am not aware of a 
> suitable alternate forum. If this does not interest you, please simply ignore 
> - I already know that this may be somewhat OT.
> 
> Thanks.
> --
> 
> You very likely have heard of ChatGPT, the conversation interface on top of 
> the GPT-3 large language model and that it can generate code. I thought it 
> doesn't do R - I was wrong. Here is a little experiment:
> 
> MY PROMPT:
> ==
> Please write me some R code that will read a tsv file, remove all rows that 
> have duplicates in the column "names", and store the median values of the 
> remaining rows in a variable called "Medians". Of course, that calculation 
> should only be done with columns that are numeric. Print the last variable in 
> order from largest to smallest.
> 
> 
> AI RESPONSE:
> 
> 
> Here is some R code that should accomplish what you've described:
> 
> (Note: the code is formatted in a little terminal window, complete with a 
> button to "copy code" at the top.)
> 
> 
> # Read the tsv file
> data <- read.table("file.tsv", header = TRUE, sep = "\t")
> 
> # Remove rows with duplicates in the "names" column
> data <- data[!duplicated(data$names), ]
> 
> # Select only the numeric columns
> num_columns <- sapply(data, is.numeric)
> numeric_data <- data[, num_columns]
> 
> # Calculate the medians for each column
> Medians <- apply(numeric_data, 1, median)
> 
> # Sort Medians in descending order
> Medians <- sort(Medians, decreasing = TRUE)
> 
> # Print Medians
> print(Medians)
> 
> 
> This code assumes that the tsv file is called "file.tsv" and is stored in the 
> current working directory. It also assumes that the first row of the file 
> contains the column names.
> 
> 
> 
> MY PROMPT:
> ==
> Can you provide me with a bit of sample data that I can use to test this code?
> 
> AI RESPONSE:
> 
> Certainly! Here is some sample data that you can use to test the code I 
> provided:
> 
> 
> name  age height  weight
> Alice 23  5.6 120
> Bob   32  6.2 180
> Charlie   29  5.9 160
> Alice 23  5.6 120
> David 25  5.8 140
> Eve   27  5.5 110
> 
> 
> You can save this data as a tsv file and use it to test the code.
> 
> 
> --
> 
> Notes: 
> 1) The code, as written does not work with the test data. The reason is that 
> the test data specifies a column "name", but the code assumes "names" (which 
> I specified in the instructions). Once this is fixed in the test data, the 
> code executes.
> 
> 2) The instructions are not accurately translated: I had asked for row means. 
> But changing the margin in the apply() statement from 2 to 1 does the correct 
> thing.
> 
> 3) Note that the test data contains both numeric and non-numeric columns. 
> Also, the name "Alice" is duplicated in the test data, which is subtle, and 
> the right thing to do.
> 
> 
> --
> 
> A similar experiment, run on the beta-site of OpenAI's Codex code-completion 
> tool gives the following code (complete with the comments.):
> 
> 
> # Read in the data
> data <- read.table("mydata.tsv", sep = "\t", header = TRUE)
> 
> # Remove all rows that have duplicates in the column "names"
> unique_names <- unique(data$names)
> data_unique <- data[data$names %in% unique_names,]
> 
> # Calculate medians 
> Medians <- sapply(data_unique[, sapply(data_unique, is.numeric)], median, 
> na.rm = TRUE)
> 
> # Print last variable in order form largest to smallest
> rev(sort(Medians[length(Medians)]))
> 
> 
> Note that the strategy is quite different (e.g using %in%, not duplicated() 
> ), the interpretation of "last variable" is technically correct but not what 
> I had in mind (ChatGPT got that right though).
> 
> 
> Changing my prompts slightly resulted it going for a dplyr solution instead, 
> complete with %>% idioms etc ... again, syntactically correct but not giving 
> me the fully correct results. 
> 
> 

Re: [R] Amazing AI

[Boris Steipe]

Exactly. But not just "error prone", rather: eloquently and confidently 
incorrect. And that in itself is a problem. When I evaluate students' work, I 
implicitly do so from a mental model of the student - aptitude, ability, 
experience, language skills etc. That's useful for summative assessment, since 
it helps efficiency - but that won't work anymore. I see a need to assess much 
more carefully, require fine-grained referencing, check every single fact ... 
and that won't scale. And then there is also the spectre of having to decide 
when this crosses the line to "concoction" - i.e. an actual academic offence ...

Best,
Boris



> On 2022-12-19, at 03:58, Milan Glacier  wrote:
> 
> [You don't often get email from n...@milanglacier.com. Learn why this is 
> important at https://aka.ms/LearnAboutSenderIdentification ]
> 
> On 12/18/22 19:01, Boris Steipe wrote:
>> Technically not a help question. But crucial to be aware of, especially for 
>> those of us in academia, or otherwise teaching R. I am not aware of a 
>> suitable alternate forum. If this does not interest you, please simply 
>> ignore - I already know that this may be somewhat OT.
>> 
>> Thanks.
>> --
>> 
>> You very likely have heard of ChatGPT, the conversation interface on top of 
>> the GPT-3 large language model and that it can generate code. I thought it 
>> doesn't do R - I was wrong. Here is a little experiment:
>> Note that the strategy is quite different (e.g using %in%, not duplicated() 
>> ), the interpretation of "last variable" is technically correct but not what 
>> I had in mind (ChatGPT got that right though).
>> 
>> 
>> Changing my prompts slightly resulted it going for a dplyr solution instead, 
>> complete with %>% idioms etc ... again, syntactically correct but not giving 
>> me the fully correct results.
>> 
>> --
>> 
>> Bottom line: The AI's ability to translate natural language instructions 
>> into code is astounding. Errors the AI makes are subtle and probably not 
>> easy to fix if you don't already know what you are doing. But the way that 
>> this can be "confidently incorrect" and plausible makes it nearly impossible 
>> to detect unless you actually run the code (you may have noticed that when 
>> you read the code).
>> 
>> Will our students use it? Absolutely.
>> 
>> Will they successfully cheat with it? That depends on the assignment. We 
>> probably need to _encourage_ them to use it rather than sanction - but 
>> require them to attribute the AI, document prompts, and identify their own, 
>> additional contributions.
>> 
>> Will it help them learn? When you are aware of the issues, it may be quite 
>> useful. It may be especially useful to teach them to specify their code 
>> carefully and completely, and to ask questions in the right way. Test cases 
>> are crucial.
>> 
>> How will it affect what we do as instructors? I don't know. Really.
>> 
>> And the future? I am not pleased to extrapolate to a job market in which 
>> they compete with knowledge workers who work 24/7 without benefits, vacation 
>> pay, or even a salary. They'll need to rethink the value of their investment 
>> in an academic education. We'll need to rethink what we do to provide value 
>> above and beyond what AI's can do. (Nb. all of the arguments I hear about 
>> why humans will always be better etc. are easily debunked, but that's even 
>> more OT :-)
>> 
>> 
>> 
>> If you have thoughts to share how your institution is thinking about 
>> academic integrity in this situation, or creative ideas how to integrate 
>> this into teaching, I'd love to hear from you.
> 
> *NEVER* let the AI misleading the students! ChatGPT gives you seemingly
> sound but actually *wrong* code!
> 
> ChatGPT never understands the formal abstraction behind the code, it
> just understands the shallow text pattern (and the syntax rules) in the
> code. And it often gives you the code that seemingly correct but indeed
> wrongly output. If it is used with code completion, then it is okay
> (just like github copilot), since the coder need to modify the code
> after getting the completion. But if you want to use ChatGPT for
> students to query information / writing code, it is error proning!

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Amazing AI

[Boris Steipe]

Yes ... but that "1" is my correction to the AI's suggested "2".
Sorry if that was confusing.

:-)



> On 2022-12-19, at 14:10, John Kane  wrote:
> 
> Does not Medians <- apply(numeric_data, 1, median) give us the rom medians?
> 
> On Mon, 19 Dec 2022 at 05:52, Milan Glacier  wrote:
> On 12/18/22 19:01, Boris Steipe wrote:
> >Technically not a help question. But crucial to be aware of, especially for 
> >those of us in academia, or otherwise teaching R. I am not aware of a 
> >suitable alternate forum. If this does not interest you, please simply 
> >ignore - I already know that this may be somewhat OT.
> >
> >Thanks.
> >--
> >
> >You very likely have heard of ChatGPT, the conversation interface on top of 
> >the GPT-3 large language model and that it can generate code. I thought it 
> >doesn't do R - I was wrong. Here is a little experiment:
> >Note that the strategy is quite different (e.g using %in%, not duplicated() 
> >), the interpretation of "last variable" is technically correct but not what 
> >I had in mind (ChatGPT got that right though).
> >
> >
> >Changing my prompts slightly resulted it going for a dplyr solution instead, 
> >complete with %>% idioms etc ... again, syntactically correct but not giving 
> >me the fully correct results.
> >
> >--
> >
> >Bottom line: The AI's ability to translate natural language instructions 
> >into code is astounding. Errors the AI makes are subtle and probably not 
> >easy to fix if you don't already know what you are doing. But the way that 
> >this can be "confidently incorrect" and plausible makes it nearly impossible 
> >to detect unless you actually run the code (you may have noticed that when 
> >you read the code).
> >
> >Will our students use it? Absolutely.
> >
> >Will they successfully cheat with it? That depends on the assignment. We 
> >probably need to _encourage_ them to use it rather than sanction - but 
> >require them to attribute the AI, document prompts, and identify their own, 
> >additional contributions.
> >
> >Will it help them learn? When you are aware of the issues, it may be quite 
> >useful. It may be especially useful to teach them to specify their code 
> >carefully and completely, and to ask questions in the right way. Test cases 
> >are crucial.
> >
> >How will it affect what we do as instructors? I don't know. Really.
> >
> >And the future? I am not pleased to extrapolate to a job market in which 
> >they compete with knowledge workers who work 24/7 without benefits, vacation 
> >pay, or even a salary. They'll need to rethink the value of their investment 
> >in an academic education. We'll need to rethink what we do to provide value 
> >above and beyond what AI's can do. (Nb. all of the arguments I hear about 
> >why humans will always be better etc. are easily debunked, but that's even 
> >more OT :-)
> >
> >
> >
> >If you have thoughts to share how your institution is thinking about 
> >academic integrity in this situation, or creative ideas how to integrate 
> >this into teaching, I'd love to hear from you.
> 
> *NEVER* let the AI misleading the students! ChatGPT gives you seemingly
> sound but actually *wrong* code!
> 
> ChatGPT never understands the formal abstraction behind the code, it
> just understands the shallow text pattern (and the syntax rules) in the
> code. And it often gives you the code that seemingly correct but indeed
> wrongly output. If it is used with code completion, then it is okay
> (just like github copilot), since the coder need to modify the code
> after getting the completion. But if you want to use ChatGPT for
> students to query information / writing code, it is error proning!
> 
> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 
> 
> -- 
> John Kane
> Kingston ON Canada

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Amazing AI

[John Kane]

Does not Medians <- apply(numeric_data, 1, median) give us the rom medians?

On Mon, 19 Dec 2022 at 05:52, Milan Glacier  wrote:

> On 12/18/22 19:01, Boris Steipe wrote:
> >Technically not a help question. But crucial to be aware of, especially
> for those of us in academia, or otherwise teaching R. I am not aware of a
> suitable alternate forum. If this does not interest you, please simply
> ignore - I already know that this may be somewhat OT.
> >
> >Thanks.
> >--
> >
> >You very likely have heard of ChatGPT, the conversation interface on top
> of the GPT-3 large language model and that it can generate code. I thought
> it doesn't do R - I was wrong. Here is a little experiment:
> >Note that the strategy is quite different (e.g using %in%, not
> duplicated() ), the interpretation of "last variable" is technically
> correct but not what I had in mind (ChatGPT got that right though).
> >
> >
> >Changing my prompts slightly resulted it going for a dplyr solution
> instead, complete with %>% idioms etc ... again, syntactically correct but
> not giving me the fully correct results.
> >
> >--
> >
> >Bottom line: The AI's ability to translate natural language instructions
> into code is astounding. Errors the AI makes are subtle and probably not
> easy to fix if you don't already know what you are doing. But the way that
> this can be "confidently incorrect" and plausible makes it nearly
> impossible to detect unless you actually run the code (you may have noticed
> that when you read the code).
> >
> >Will our students use it? Absolutely.
> >
> >Will they successfully cheat with it? That depends on the assignment. We
> probably need to _encourage_ them to use it rather than sanction - but
> require them to attribute the AI, document prompts, and identify their own,
> additional contributions.
> >
> >Will it help them learn? When you are aware of the issues, it may be
> quite useful. It may be especially useful to teach them to specify their
> code carefully and completely, and to ask questions in the right way. Test
> cases are crucial.
> >
> >How will it affect what we do as instructors? I don't know. Really.
> >
> >And the future? I am not pleased to extrapolate to a job market in which
> they compete with knowledge workers who work 24/7 without benefits,
> vacation pay, or even a salary. They'll need to rethink the value of their
> investment in an academic education. We'll need to rethink what we do to
> provide value above and beyond what AI's can do. (Nb. all of the arguments
> I hear about why humans will always be better etc. are easily debunked, but
> that's even more OT :-)
> >
> >
> >
> >If you have thoughts to share how your institution is thinking about
> academic integrity in this situation, or creative ideas how to integrate
> this into teaching, I'd love to hear from you.
>
> *NEVER* let the AI misleading the students! ChatGPT gives you seemingly
> sound but actually *wrong* code!
>
> ChatGPT never understands the formal abstraction behind the code, it
> just understands the shallow text pattern (and the syntax rules) in the
> code. And it often gives you the code that seemingly correct but indeed
> wrongly output. If it is used with code completion, then it is okay
> (just like github copilot), since the coder need to modify the code
> after getting the completion. But if you want to use ChatGPT for
> students to query information / writing code, it is error proning!
>
> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
John Kane
Kingston ON Canada

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Integer division

[Bert Gunter]

"It would probably be helpful to add a short paragraph to  ?Arithmetic
about the fact that R's %% uses the "floored" version, as
recommended by Donald Knuth and as documented on the above
Wikipedia page."

Agreed (who am I to disagree?!).

But perhaps something simpler like:

"a %% b must always be between 0 (inclusive) and b(exclusive)."

ergo, for examples,
> 5 %% 3
[1] 2  ## 5 = 1*3 + 2
> 5 %% -3
[1] -1 ## 5 = 2*(-3) + (-1)
> -5 %% 3
[1] 1  ## -5 = (-2) * 3 + 1
> -5 %% -3
[1] -2  ## -5 = 1* (-3) + (-2)

My point is that I believe it is easier for most non-math folks to groc the
relationship through a simple rule and such examples than via an algorithm.

Feel free to disagree, as this is just my opinion.

Cheers,
Bert




On Mon, Dec 19, 2022 at 9:30 AM Martin Maechler 
wrote:

> > Jeff Newmiller
> > on Mon, 19 Dec 2022 08:37:32 -0800 writes:
>
> > See https://en.m.wikipedia.org/wiki/Modulo_operation,
> > Variants of the definition, esp the point that Knuth
> > recommended the floor definition. The behavior of %/%
> > follows from the definition of %% given the documented
> > relation in ?Arithmetic.
>
> > R is not obligated to repeat the mistakes of C or Fortran.
>
> Fortune nomination!  ==> BCC:  maintainer("fortunes")
>
> The Wikipedia page is indeed revealing amazing facts about how
> differently this has been implemented in computer languages.
>
> Then, after all,  Göran still got a point to make here, given that
> not anymore are all R users equipped with a Ph.D in math or equivalent..:
>
> It would probably be helpful to add a short paragraph to  ?Arithmetic
> about the fact that R's %% uses the "floored" version, as
> recommended by Donald Knuth and as documented on the above
> Wikipedia page.
>
> Martin
>
> > On December 19, 2022 7:15:01 AM PST, "Göran Broström"
> >  wrote:
> >>
> >>
> >> Den 2022-12-19 kl. 15:41, skrev Martin Maechler:
>  Göran Broström on Mon, 19 Dec 2022 14:22:00 +0100
>  writes:
> >>>
> >>> > I have a long vector x with five-digit codes where the
> >>> > first digit of each is of special interest, so I
> >>> extracted > them through
> >>>
> >>> >> y <- x %/% 1
> >>>
> >>> > but to my surprise y contained the value -1 in some >
> >>> places. It turned out that x contains -1 as a symbol for
> >>> > 'missing value' so in effect I found that
> >>>
> >>> >> -1 %/% 1 == -1
> >>>
> >>> > Had to check the help page for "%/%", and the first >
> >>> relevant comment I found was:
> >>>
> >>> > "Users are sometimes surprised by the value returned".
> >>>
> >>> > No surprise there. Further down:
> >>>
> >>> > ‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’ >
> >>> indicates integer division.  It is guaranteed that
> >>>
> >>> > ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding >
> >>> error)
> >>>
> >>> > I did expect (a %/% b) to return round(a / b), like >
> >>> gfortran and gcc,
> >>>
> >>> What???  I cannot believe you.
> >>
> >> Well, you shouldn't, I generalized too far.
> >>>
> >>> No time for checking now, but I bet that 8 / 3 gives 2
> >>> and not 3 in C and Fortran (and hence gcc, etc)
> >>
> >> But compare -8 %/% 3 in R and -8 / 3 in C/Fortran.
> >>
> >> G,
> >>
> >>>
> >>>
> >>> > but instead I get floor(a / b) in > R. What is the
> >>> reason for these different definitions? And > shouldn't
> >>> R's definition be documented?
> >>>
> >>>
> >>>
> >>> > Thanks, Göran
> >>>
> >>> > __ >
> >>> R-help@r-project.org mailing list -- To UNSUBSCRIBE and
> >>> > more, see https://stat.ethz.ch/mailman/listinfo/r-help
> >>> > PLEASE do read the posting guide >
> >>> http://www.R-project.org/posting-guide.html and provide
> >>> > commented, minimal, self-contained, reproducible code.
> >>
> >> __
> >> R-help@r-project.org mailing list -- To UNSUBSCRIBE and
> >> more, see https://stat.ethz.ch/mailman/listinfo/r-help
> >> PLEASE do read the posting guide
> >> http://www.R-project.org/posting-guide.html and provide
> >> commented, minimal, self-contained, reproducible code.
>
> > --
> > Sent from my phone. Please excuse my brevity.
>
> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read 

Re: [R] Integer division

[Martin Maechler]

> Jeff Newmiller 
> on Mon, 19 Dec 2022 08:37:32 -0800 writes:

> See https://en.m.wikipedia.org/wiki/Modulo_operation,
> Variants of the definition, esp the point that Knuth
> recommended the floor definition. The behavior of %/%
> follows from the definition of %% given the documented
> relation in ?Arithmetic.

> R is not obligated to repeat the mistakes of C or Fortran.

Fortune nomination!  ==> BCC:  maintainer("fortunes")

The Wikipedia page is indeed revealing amazing facts about how
differently this has been implemented in computer languages.

Then, after all,  Göran still got a point to make here, given that
not anymore are all R users equipped with a Ph.D in math or equivalent..:

It would probably be helpful to add a short paragraph to  ?Arithmetic
about the fact that R's %% uses the "floored" version, as
recommended by Donald Knuth and as documented on the above
Wikipedia page.

Martin

> On December 19, 2022 7:15:01 AM PST, "Göran Broström"
>  wrote:
>> 
>> 
>> Den 2022-12-19 kl. 15:41, skrev Martin Maechler:
 Göran Broström on Mon, 19 Dec 2022 14:22:00 +0100
 writes:
>>> 
>>> > I have a long vector x with five-digit codes where the
>>> > first digit of each is of special interest, so I
>>> extracted > them through
>>> 
>>> >> y <- x %/% 1
>>> 
>>> > but to my surprise y contained the value -1 in some >
>>> places. It turned out that x contains -1 as a symbol for
>>> > 'missing value' so in effect I found that
>>> 
>>> >> -1 %/% 1 == -1
>>> 
>>> > Had to check the help page for "%/%", and the first >
>>> relevant comment I found was:
>>> 
>>> > "Users are sometimes surprised by the value returned".
>>> 
>>> > No surprise there. Further down:
>>> 
>>> > ‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’ >
>>> indicates integer division.  It is guaranteed that
>>> 
>>> > ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding >
>>> error)
>>> 
>>> > I did expect (a %/% b) to return round(a / b), like >
>>> gfortran and gcc,
>>> 
>>> What???  I cannot believe you.
>> 
>> Well, you shouldn't, I generalized too far.
>>> 
>>> No time for checking now, but I bet that 8 / 3 gives 2
>>> and not 3 in C and Fortran (and hence gcc, etc)
>> 
>> But compare -8 %/% 3 in R and -8 / 3 in C/Fortran.
>> 
>> G,
>> 
>>> 
>>> 
>>> > but instead I get floor(a / b) in > R. What is the
>>> reason for these different definitions? And > shouldn't
>>> R's definition be documented?
>>> 
>>> 
>>> 
>>> > Thanks, Göran
>>> 
>>> > __ >
>>> R-help@r-project.org mailing list -- To UNSUBSCRIBE and
>>> > more, see https://stat.ethz.ch/mailman/listinfo/r-help
>>> > PLEASE do read the posting guide >
>>> http://www.R-project.org/posting-guide.html and provide
>>> > commented, minimal, self-contained, reproducible code.
>> 
>> __
>> R-help@r-project.org mailing list -- To UNSUBSCRIBE and
>> more, see https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html and provide
>> commented, minimal, self-contained, reproducible code.

> -- 
> Sent from my phone. Please excuse my brevity.

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Integer division

[Jeff Newmiller]

See https://en.m.wikipedia.org/wiki/Modulo_operation, Variants of the 
definition, esp the point that Knuth recommended the floor definition. The 
behavior of %/% follows from the definition of %% given the documented relation 
in ?Arithmetic.

R is not obligated to repeat the mistakes of C or Fortran.

On December 19, 2022 7:15:01 AM PST, "Göran Broström"  wrote:
>
>
>Den 2022-12-19 kl. 15:41, skrev Martin Maechler:
>>> Göran Broström
>>>  on Mon, 19 Dec 2022 14:22:00 +0100 writes:
>> 
>>  > I have a long vector x with five-digit codes where the
>>  > first digit of each is of special interest, so I extracted
>>  > them through
>> 
>>  >> y <- x %/% 1
>> 
>>  > but to my surprise y contained the value -1 in some
>>  > places. It turned out that x contains -1 as a symbol for
>>  > 'missing value' so in effect I found that
>> 
>>  >> -1 %/% 1 == -1
>> 
>>  > Had to check the help page for "%/%", and the first
>>  > relevant comment I found was:
>> 
>>  > "Users are sometimes surprised by the value returned".
>> 
>>  > No surprise there. Further down:
>> 
>>  > ‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’
>>  > indicates integer division.  It is guaranteed that
>> 
>>  >   ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding
>>  > error)
>> 
>>  > I did expect (a %/% b) to return round(a / b), like
>>  > gfortran and gcc,
>> 
>> What???  I cannot believe you.
>
>Well, you shouldn't, I generalized too far.
>> 
>> No time for checking now, but I bet that
>> 8 / 3  gives 2 and not 3  in C and Fortran
>> (and hence gcc, etc)
>
>But compare -8 %/% 3 in R and -8 / 3 in C/Fortran.
>
>G,
>
>> 
>> 
>>  > but instead I get floor(a / b) in
>>  > R. What is the reason for these different definitions? And
>>  > shouldn't R's definition be documented?
>> 
>> 
>> 
>>  > Thanks, Göran
>> 
>>  > __
>>  > R-help@r-project.org mailing list -- To UNSUBSCRIBE and
>>  > more, see https://stat.ethz.ch/mailman/listinfo/r-help
>>  > PLEASE do read the posting guide
>>  > http://www.R-project.org/posting-guide.html and provide
>>  > commented, minimal, self-contained, reproducible code.
>
>__
>R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>and provide commented, minimal, self-contained, reproducible code.

-- 
Sent from my phone. Please excuse my brevity.

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Integer division

[Bert Gunter]

If R does exactly what it says it does, why are you surprised, whether or
not that is what other languages do?  (Please excuse my fractured English).
(Note: -8 = 1 +  3*(-3)  =  (-8 %% 3)  + 3 * (-8 %/% 3 ), exactly as the
Help excerpt you cited says)

Of course, one may always question the wisdom of whatever policy R chooses,
but that gets into the bowels of what standards to follow, corner cases,
and all that, about which I have nothing to say. My only  point is the
above.



-- Bert

On Mon, Dec 19, 2022 at 7:15 AM Göran Broström  wrote:

>
>
> Den 2022-12-19 kl. 15:41, skrev Martin Maechler:
> >> Göran Broström
> >>  on Mon, 19 Dec 2022 14:22:00 +0100 writes:
> >
> >  > I have a long vector x with five-digit codes where the
> >  > first digit of each is of special interest, so I extracted
> >  > them through
> >
> >  >> y <- x %/% 1
> >
> >  > but to my surprise y contained the value -1 in some
> >  > places. It turned out that x contains -1 as a symbol for
> >  > 'missing value' so in effect I found that
> >
> >  >> -1 %/% 1 == -1
> >
> >  > Had to check the help page for "%/%", and the first
> >  > relevant comment I found was:
> >
> >  > "Users are sometimes surprised by the value returned".
> >
> >  > No surprise there. Further down:
> >
> >  > ‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’
> >  > indicates integer division.  It is guaranteed that
> >
> >  >   ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding
> >  > error)
> >
> >  > I did expect (a %/% b) to return round(a / b), like
> >  > gfortran and gcc,
> >
> > What???  I cannot believe you.
>
> Well, you shouldn't, I generalized too far.
> >
> > No time for checking now, but I bet that
> > 8 / 3  gives 2 and not 3  in C and Fortran
> > (and hence gcc, etc)
>
> But compare -8 %/% 3 in R and -8 / 3 in C/Fortran.
>
> G,
>
> >
> >
> >  > but instead I get floor(a / b) in
> >  > R. What is the reason for these different definitions? And
> >  > shouldn't R's definition be documented?
> >
> >
> >
> >  > Thanks, Göran
> >
> >  > __
> >  > R-help@r-project.org mailing list -- To UNSUBSCRIBE and
> >  > more, see https://stat.ethz.ch/mailman/listinfo/r-help
> >  > PLEASE do read the posting guide
> >  > http://www.R-project.org/posting-guide.html and provide
> >  > commented, minimal, self-contained, reproducible code.
>
> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Integer division

[Göran Broström]

Den 2022-12-19 kl. 15:41, skrev Martin Maechler:

Göran Broström
 on Mon, 19 Dec 2022 14:22:00 +0100 writes:


 > I have a long vector x with five-digit codes where the
 > first digit of each is of special interest, so I extracted
 > them through

 >> y <- x %/% 1

 > but to my surprise y contained the value -1 in some
 > places. It turned out that x contains -1 as a symbol for
 > 'missing value' so in effect I found that

 >> -1 %/% 1 == -1

 > Had to check the help page for "%/%", and the first
 > relevant comment I found was:

 > "Users are sometimes surprised by the value returned".

 > No surprise there. Further down:

 > ‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’
 > indicates integer division.  It is guaranteed that

 >   ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding
 > error)

 > I did expect (a %/% b) to return round(a / b), like
 > gfortran and gcc,

What???  I cannot believe you.


Well, you shouldn't, I generalized too far.


No time for checking now, but I bet that
8 / 3  gives 2 and not 3  in C and Fortran
(and hence gcc, etc)


But compare -8 %/% 3 in R and -8 / 3 in C/Fortran.

G,




 > but instead I get floor(a / b) in
 > R. What is the reason for these different definitions? And
 > shouldn't R's definition be documented?



 > Thanks, Göran

 > __
 > R-help@r-project.org mailing list -- To UNSUBSCRIBE and
 > more, see https://stat.ethz.ch/mailman/listinfo/r-help
 > PLEASE do read the posting guide
 > http://www.R-project.org/posting-guide.html and provide
 > commented, minimal, self-contained, reproducible code.


__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Integer division

[Martin Maechler]

> Göran Broström 
> on Mon, 19 Dec 2022 14:22:00 +0100 writes:

> I have a long vector x with five-digit codes where the
> first digit of each is of special interest, so I extracted
> them through

>> y <- x %/% 1

> but to my surprise y contained the value -1 in some
> places. It turned out that x contains -1 as a symbol for
> 'missing value' so in effect I found that

>> -1 %/% 1 == -1

> Had to check the help page for "%/%", and the first
> relevant comment I found was:

> "Users are sometimes surprised by the value returned".

> No surprise there. Further down:

> ‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’
> indicates integer division.  It is guaranteed that

>   ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding
> error)

> I did expect (a %/% b) to return round(a / b), like
> gfortran and gcc, 

What???  I cannot believe you.

No time for checking now, but I bet that 
8 / 3  gives 2 and not 3  in C and Fortran
(and hence gcc, etc) 


> but instead I get floor(a / b) in
> R. What is the reason for these different definitions? And
> shouldn't R's definition be documented?



> Thanks, Göran

> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and
> more, see https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html and provide
> commented, minimal, self-contained, reproducible code.

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

[R] Integer division

[Göran Broström]

I have a long vector x with five-digit codes where the first digit of 
each is of special interest, so I extracted them through


> y <- x %/% 1

but to my surprise y contained the value -1 in some places. It turned 
out that x contains -1 as a symbol for 'missing value' so in effect I 
found that


> -1 %/% 1 == -1

Had to check the help page for "%/%", and the first relevant comment I 
found was:


"Users are sometimes surprised by the value returned".

No surprise there. Further down:

‘%%’ indicates ‘x mod y’ (“x modulo y”) and ‘%/%’ indicates
 integer division.  It is guaranteed that

 ‘ x == (x %% y) + y * (x %/% y) ’ (up to rounding error)

I did expect  (a %/% b) to return round(a / b), like gfortran and gcc, 
but instead I get floor(a / b) in R. What is the reason for these 
different definitions? And shouldn't R's definition be documented?


Thanks, Göran

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] [Rd] I do not want that R CMD build removes temp directory

[Serguei Sokol via R-help]

Le 19/12/2022 à 10:52, Witold E Wolski a écrit :

Dear Uwe,

Unfortunately there isn't much of an output. This is all what I have:

$ R CMD INSTALL --log prolfqua
Warning: unknown option '--log'
* installing to library 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2'
* installing *source* package 'prolfqua' ...
** using staged installation
** R
** data
** inst
** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/prolfqua'

Also with --no-test-load option the install is failing :

$ R CMD INSTALL --clean --no-test-load prolfqua

* installing to library 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2'
* installing *source* package 'prolfqua' ...
** using staged installation
** R
** data
** inst
** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/prolfqua'

And including "--no-clean-on-error" also does not help because the
installation directory is empty.
You don't show the full command you run. If you included also "--clean" 
as in previous command, may be it just does what is asked, i.e. clean 
the temp dir ?
I don't know what prevails: "--clean" or "--no-clean-on-error" if put 
simultaneously.


Best,
Serguei.




Tested the install, on macos M1, linux ARM64, linux x86, Windows 64,
and it works everywhere except
Parallels Windows 64 on ARM M1.

R version 4.2.2 (2022-10-31 ucrt) -- "Innocent and Trusting"
Copyright (C) 2022 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)

best regards
Witek



On Fri, 16 Dec 2022 at 11:24, Uwe Ligges
 wrote:



On 15.12.2022 21:47, Witold E Wolski wrote:

Thank you Simon,

It seems not to be related to the R package but rather to the OS,
(just got the same error when installing the shiny R package from
CRAN).
I am on an M1 mac running Windows ARM in Parallels. Installed a
x86_64-w64 R version.



"** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'shiny'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/shiny'
Warning in install.packages :"

Can we please have the full output?

Best,
Uwe Ligges




On Thu, 15 Dec 2022 at 19:09, Simon Urbanek  wrote:

Yes:

$ R CMD INSTALL --help | grep error
--no-clean-on-error   do not remove installed package on error

But probably more commonly used way is to install the package from its unpacked 
directory as that avoids the use of temporary directories in the first place.

In you case you can also use --no-test-load and the non-functional package will 
still be installed so you can inspect it.

Cheers,
Simon

PS: please don't cross-post



On Dec 16, 2022, at 7:01 AM, Witold E Wolski  wrote:

I am getting a package build error, and can not figure out the problem.
The error is
"
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/
"
However since R CMD build removes the temp directory and does not give
any other errors how can I find out what the build problem is?

Is there a way to disable the temp directory removal?

Best Regards
Witek
--
Witold Eryk Wolski

__
r-de...@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel






--
Witold Eryk Wolski

__
r-de...@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel



--
Serguei Sokol
Ingenieur de recherche INRAE

Cellule Mathématiques
TBI, INSA/INRAE UMR 792, INSA/CNRS UMR 5504
135 Avenue de Rangueil
31077 Toulouse Cedex 04

tel: +33 5 61 55 98 49
email: so...@insa-toulouse.fr
http://www.toulouse-biotechnology-institute.fr/en/technology_platforms/mathematics-cell.html

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Amazing AI

[Milan Glacier]

On 12/18/22 19:01, Boris Steipe wrote:

Technically not a help question. But crucial to be aware of, especially for 
those of us in academia, or otherwise teaching R. I am not aware of a suitable 
alternate forum. If this does not interest you, please simply ignore - I 
already know that this may be somewhat OT.

Thanks.
--

You very likely have heard of ChatGPT, the conversation interface on top of the 
GPT-3 large language model and that it can generate code. I thought it doesn't 
do R - I was wrong. Here is a little experiment:
Note that the strategy is quite different (e.g using %in%, not duplicated() ), the 
interpretation of "last variable" is technically correct but not what I had in 
mind (ChatGPT got that right though).


Changing my prompts slightly resulted it going for a dplyr solution instead, 
complete with %>% idioms etc ... again, syntactically correct but not giving me 
the fully correct results.

--

Bottom line: The AI's ability to translate natural language instructions into code is 
astounding. Errors the AI makes are subtle and probably not easy to fix if you don't 
already know what you are doing. But the way that this can be "confidently 
incorrect" and plausible makes it nearly impossible to detect unless you actually 
run the code (you may have noticed that when you read the code).

Will our students use it? Absolutely.

Will they successfully cheat with it? That depends on the assignment. We 
probably need to _encourage_ them to use it rather than sanction - but require 
them to attribute the AI, document prompts, and identify their own, additional 
contributions.

Will it help them learn? When you are aware of the issues, it may be quite 
useful. It may be especially useful to teach them to specify their code 
carefully and completely, and to ask questions in the right way. Test cases are 
crucial.

How will it affect what we do as instructors? I don't know. Really.

And the future? I am not pleased to extrapolate to a job market in which they 
compete with knowledge workers who work 24/7 without benefits, vacation pay, or 
even a salary. They'll need to rethink the value of their investment in an 
academic education. We'll need to rethink what we do to provide value above and 
beyond what AI's can do. (Nb. all of the arguments I hear about why humans will 
always be better etc. are easily debunked, but that's even more OT :-)



If you have thoughts to share how your institution is thinking about academic 
integrity in this situation, or creative ideas how to integrate this into 
teaching, I'd love to hear from you.


*NEVER* let the AI misleading the students! ChatGPT gives you seemingly
sound but actually *wrong* code!

ChatGPT never understands the formal abstraction behind the code, it
just understands the shallow text pattern (and the syntax rules) in the
code. And it often gives you the code that seemingly correct but indeed
wrongly output. If it is used with code completion, then it is okay
(just like github copilot), since the coder need to modify the code
after getting the completion. But if you want to use ChatGPT for
students to query information / writing code, it is error proning!

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] [Rd] I do not want that R CMD build removes temp directory

[Uwe Ligges]

Ah, before that you wrote

""
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:Users
"

and C:Users  without a slash sonds suspicious. Now with the new output I 
do not see where the issue is from. Does this also happen when you try 
it on winbuilder?


Best,
Uwe Ligges

On 19.12.2022 10:52, Witold E Wolski wrote:

Dear Uwe,

Unfortunately there isn't much of an output. This is all what I have:

$ R CMD INSTALL --log prolfqua
Warning: unknown option '--log'
* installing to library 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2'
* installing *source* package 'prolfqua' ...
** using staged installation
** R
** data
** inst
** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/prolfqua'

Also with --no-test-load option the install is failing :

$ R CMD INSTALL --clean --no-test-load prolfqua

* installing to library 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2'
* installing *source* package 'prolfqua' ...
** using staged installation
** R
** data
** inst
** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/prolfqua'

And including "--no-clean-on-error" also does not help because the
installation directory is empty.


Tested the install, on macos M1, linux ARM64, linux x86, Windows 64,
and it works everywhere except
Parallels Windows 64 on ARM M1.

R version 4.2.2 (2022-10-31 ucrt) -- "Innocent and Trusting"
Copyright (C) 2022 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)

best regards
Witek



On Fri, 16 Dec 2022 at 11:24, Uwe Ligges
 wrote:




On 15.12.2022 21:47, Witold E Wolski wrote:

Thank you Simon,

It seems not to be related to the R package but rather to the OS,
(just got the same error when installing the shiny R package from
CRAN).
I am on an M1 mac running Windows ARM in Parallels. Installed a
x86_64-w64 R version.



"** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'shiny'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/shiny'
Warning in install.packages :"


Can we please have the full output?

Best,
Uwe Ligges




On Thu, 15 Dec 2022 at 19:09, Simon Urbanek  wrote:


Yes:

$ R CMD INSTALL --help | grep error
--no-clean-on-error   do not remove installed package on error

But probably more commonly used way is to install the package from its unpacked 
directory as that avoids the use of temporary directories in the first place.

In you case you can also use --no-test-load and the non-functional package will 
still be installed so you can inspect it.

Cheers,
Simon

PS: please don't cross-post



On Dec 16, 2022, at 7:01 AM, Witold E Wolski  wrote:

I am getting a package build error, and can not figure out the problem.
The error is
"
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/
"
However since R CMD build removes the temp directory and does not give
any other errors how can I find out what the build problem is?

Is there a way to disable the temp directory removal?

Best Regards
Witek
--
Witold Eryk Wolski

__
r-de...@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel










--
Witold Eryk Wolski

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] [Rd] I do not want that R CMD build removes temp directory

[Witold E Wolski]

Dear Uwe,

Unfortunately there isn't much of an output. This is all what I have:

$ R CMD INSTALL --log prolfqua
Warning: unknown option '--log'
* installing to library 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2'
* installing *source* package 'prolfqua' ...
** using staged installation
** R
** data
** inst
** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/prolfqua'

Also with --no-test-load option the install is failing :

$ R CMD INSTALL --clean --no-test-load prolfqua

* installing to library 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2'
* installing *source* package 'prolfqua' ...
** using staged installation
** R
** data
** inst
** byte-compile and prepare package for lazy loading
ERROR: lazy loading failed for package 'prolfqua'
* removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/prolfqua'

And including "--no-clean-on-error" also does not help because the
installation directory is empty.


Tested the install, on macos M1, linux ARM64, linux x86, Windows 64,
and it works everywhere except
Parallels Windows 64 on ARM M1.

R version 4.2.2 (2022-10-31 ucrt) -- "Innocent and Trusting"
Copyright (C) 2022 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)

best regards
Witek



On Fri, 16 Dec 2022 at 11:24, Uwe Ligges
 wrote:
>
>
>
> On 15.12.2022 21:47, Witold E Wolski wrote:
> > Thank you Simon,
> >
> > It seems not to be related to the R package but rather to the OS,
> > (just got the same error when installing the shiny R package from
> > CRAN).
> > I am on an M1 mac running Windows ARM in Parallels. Installed a
> > x86_64-w64 R version.
> >
> >
> >
> > "** byte-compile and prepare package for lazy loading
> > ERROR: lazy loading failed for package 'shiny'
> > * removing 'C:/Users/witoldwolski/AppData/Local/R/win-library/4.2/shiny'
> > Warning in install.packages :"
>
> Can we please have the full output?
>
> Best,
> Uwe Ligges
>
>
>
> > On Thu, 15 Dec 2022 at 19:09, Simon Urbanek  
> > wrote:
> >>
> >> Yes:
> >>
> >> $ R CMD INSTALL --help | grep error
> >>--no-clean-on-error   do not remove installed package on error
> >>
> >> But probably more commonly used way is to install the package from its 
> >> unpacked directory as that avoids the use of temporary directories in the 
> >> first place.
> >>
> >> In you case you can also use --no-test-load and the non-functional package 
> >> will still be installed so you can inspect it.
> >>
> >> Cheers,
> >> Simon
> >>
> >> PS: please don't cross-post
> >>
> >>
> >>> On Dec 16, 2022, at 7:01 AM, Witold E Wolski  wrote:
> >>>
> >>> I am getting a package build error, and can not figure out the problem.
> >>> The error is
> >>> "
> >>> ERROR: lazy loading failed for package 'prolfqua'
> >>> * removing 'C:/Users/
> >>> "
> >>> However since R CMD build removes the temp directory and does not give
> >>> any other errors how can I find out what the build problem is?
> >>>
> >>> Is there a way to disable the temp directory removal?
> >>>
> >>> Best Regards
> >>> Witek
> >>> --
> >>> Witold Eryk Wolski
> >>>
> >>> __
> >>> r-de...@r-project.org mailing list
> >>> https://stat.ethz.ch/mailman/listinfo/r-devel
> >>>
> >>
> >
> >



--
Witold Eryk Wolski

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Amazing AI

[Eric Berger]

Hi Boris,
I think these are good questions.
Some initial reactions:
1. with better tools available to the students, you can cover more material
at a faster pace 
2. For years, it has been possible for students to find "answers" online
(e.g. Google search). Most programmers would regard this as an essential
part of their work - looking to overcome some obstacle via searching on
StackOverflow, Google, R-Help etc*. I consider the ability to do such
searches something worth teaching. In this spirit, you could teach
effective ways to use ChatGPT.
3. Finally, as you observed, some of the ChatGPT responses have
bugs/deficiencies. For example, the Codex code does *not* remove
duplicates. It requires knowledge and skill of the material you are
teaching, for the students to be able to spot and fix the bugs. In which
case, what's your problem? 

HTH,
Eric

* and as a shout-out to Bert, they should also read the documentation -
e.g. ?subset

On Mon, Dec 19, 2022 at 12:47 AM Ebert,Timothy Aaron  wrote:

> It would help students formulate a plan for coding. Successful students
> will be able to give good directions that the AI can turn into good code.
> This skill is essential no matter who writes the program.
> In more advanced classes I might collect some data sets designed to cause
> the AI problems. Another option is to make tests where students have to
> write code on paper or multiple guess where students must choose between
> similar code snips.
>
>
> -Original Message-
> From: R-help  On Behalf Of Boris Steipe
> Sent: Sunday, December 18, 2022 2:01 PM
> To: r-help@r-project.org
> Subject: [R] Amazing AI
>
> [External Email]
>
> Technically not a help question. But crucial to be aware of, especially
> for those of us in academia, or otherwise teaching R. I am not aware of a
> suitable alternate forum. If this does not interest you, please simply
> ignore - I already know that this may be somewhat OT.
>
> Thanks.
> --
>
> You very likely have heard of ChatGPT, the conversation interface on top
> of the GPT-3 large language model and that it can generate code. I thought
> it doesn't do R - I was wrong. Here is a little experiment:
>
> MY PROMPT:
> ==
> Please write me some R code that will read a tsv file, remove all rows
> that have duplicates in the column "names", and store the median values of
> the remaining rows in a variable called "Medians". Of course, that
> calculation should only be done with columns that are numeric. Print the
> last variable in order from largest to smallest.
>
>
> AI RESPONSE:
> 
>
> Here is some R code that should accomplish what you've described:
>
> (Note: the code is formatted in a little terminal window, complete with a
> button to "copy code" at the top.)
>
> 
> # Read the tsv file
> data <- read.table("file.tsv", header = TRUE, sep = "\t")
>
> # Remove rows with duplicates in the "names" column data <-
> data[!duplicated(data$names), ]
>
> # Select only the numeric columns
> num_columns <- sapply(data, is.numeric)
> numeric_data <- data[, num_columns]
>
> # Calculate the medians for each column
> Medians <- apply(numeric_data, 1, median)
>
> # Sort Medians in descending order
> Medians <- sort(Medians, decreasing = TRUE)
>
> # Print Medians
> print(Medians)
> 
>
> This code assumes that the tsv file is called "file.tsv" and is stored in
> the current working directory. It also assumes that the first row of the
> file contains the column names.
>
>
>
> MY PROMPT:
> ==
> Can you provide me with a bit of sample data that I can use to test this
> code?
>
> AI RESPONSE:
> 
> Certainly! Here is some sample data that you can use to test the code I
> provided:
>
> 
> nameage height  weight
> Alice   23  5.6 120
> Bob 32  6.2 180
> Charlie 29  5.9 160
> Alice   23  5.6 120
> David   25  5.8 140
> Eve 27  5.5 110
> 
>
> You can save this data as a tsv file and use it to test the code.
>
>
> --
>
> Notes:
> 1) The code, as written does not work with the test data. The reason is
> that the test data specifies a column "name", but the code assumes "names"
> (which I specified in the instructions). Once this is fixed in the test
> data, the code executes.
>
> 2) The instructions are not accurately translated: I had asked for row
> means. But changing the margin in the apply() statement from 2 to 1 does
> the correct thing.
>
> 3) Note that the test data contains both numeric and non-numeric columns.
> Also, the name "Alice" is duplicated in the test data, which is subtle, and
> the right thing to do.
>
>
> --
>
> A similar experiment, run on the beta-site of OpenAI's Codex
> code-completion tool gives the following code (complete with the comments.):
>
> 
> # Read in the data
> data <-