Re: LyX+Scilab

[Jerry]

On Jan 25, 2015, at 2:56 PM, Benedict Holland benedict.m.holl...@gmail.com 
wrote:

 On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin 
 francesco.menon...@unibs.it wrote:
 Hello Benedict,
 thank you for your answer.
 I give you an example about how R is difficult for dealing with matrices: if 
 in Matlab (or Scilab or Octave or simila) I write:
 
 x(2,2)=10
 
 the following matrix is automatically created:
 
 [0 0
 0 2]
 
 while in R
 
 x[2,2]=10
 
 gives an error because I first have to declare to R that x is a matrix and, 
 furthermore, give it the right dimension.
 If, afterward, I want to add an element to the matrix x, in Matlab I simply 
 write:
 
 x(3,1)=5
 
 and I obtain the matrix:
 
 [0 0
 0 10
 5 0]
 
 while in R I cannot do that, because x has already been defined as a 2 by 2 
 matrix.
 I find all this very inefficient for dealig with matrices.
 This is my point :-)
 
 Francesco
 
I'm not familiar with R but what you are saying about R in comparison with 
Matlab or Octave is that Matlab/Octave do not check array bounds. Thus, you 
might erroneously write into a matrix element that doesn't exist and 
Matlab/Octave does not report an error. Array bounds checking is the price to 
pay for having safer code. FWIW, Python and Mathematica are examples of 
languages that are only half-safe with respect to array indexing, meaning if 
you write past the beginning of an array (usually negative indexes) it assumes 
you want to index the array backwards, from the end. These are all conveniences 
but of course can also cause erroneous results or increased time spent 
debugging. I personally am happy to write in a safe language (Ada) and spend a 
little bit more time writing code (it's easy to declare a new array and copy 
the old one into it; you can write a subroutine if you want) and virtually zero 
time chasing bugs, but others prefer otherwise and that's fine.

Jerry 

Re: LyX+Scilab

[Benedict Holland]

Naw, the issue is just that R is messy. Actually, for this particular case,
I was chuckling a bit since it isn't wrong how R did it, it is just the way
R does it. C++ is exactly the same. Unfortunately, the answer that I will
give for this thread is that the OP should really just learn R or figure
out a workaround. I have done a huge amount of work in many languages and
basically I would not wish to merge them. Loosely coupled code is much
better than tightly coupled. Reading in a CSV or a PDF is much better style
than generating it and working with it inside of a document. Eventually you
start asking yourself questions like, if I gave this Lyx document to a
friend, would they have any idea what I did and could they figure it out in
under 2 minutes? If the answer is no, I might suggest finding a different
way to attack the problem.

I was thinking about this though. For plots, R is one of the best tools you
can use. Believe you me, illustrator is probably the best but R does a
really good job and is not 800$. What I really recommend for plots is that
you save the plot and load it into the document. I have done this many
times. That way, you are regenerating the same plot every time you compile
your document. Also, as long as you can generate a PDF image, you can
include it.

Now I bet there is a case where the plot should be dynamic with every
compile and if that is the case, I don't really know where to go apart from
learning R... if only for the plotting which is some of the best you will
find anywhere. I actually find the whole idea of integrating R into Lyx to
be interesting but misses the point a bit. For tables I could see it being
very useful but within the boundaries of whatever automatically sets up the
table formatting. For images, I would save it and import it as a pdf. BTW,
I have had no end of problems working with sweave and the stata latex
output. It essentially requires me to rewrite the tables anyway. Problems
range anywhere from writing off the end of the page to generating tables
which are unreadable.

BTW, totally off topic but Python is very versatile. It actually makes much
more sense with the power behind dictionaries, document parsing, document
merging, string manipulations, data management, etc. for it to be included
in Lyx. That I am sure was a feat of engineering.

I honestly don't know what to tell the OP anymore as latex includes many
many ways to integrate PDF images within a document. Generating the PDF
outside of the document and importing it seems like far less work than
trying to get it all working in Lyx.

~Ben

On Wed, Jan 28, 2015 at 10:56 PM, Jerry lancebo...@qwest.net wrote:


 On Jan 25, 2015, at 2:56 PM, Benedict Holland 
 benedict.m.holl...@gmail.com wrote:

  On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin 
 francesco.menon...@unibs.it wrote:
  Hello Benedict,
  thank you for your answer.
  I give you an example about how R is difficult for dealing with
 matrices: if in Matlab (or Scilab or Octave or simila) I write:
 
  x(2,2)=10
 
  the following matrix is automatically created:
 
  [0 0
  0 2]
 
  while in R
 
  x[2,2]=10
 
  gives an error because I first have to declare to R that x is a matrix
 and, furthermore, give it the right dimension.
  If, afterward, I want to add an element to the matrix x, in Matlab I
 simply write:
 
  x(3,1)=5
 
  and I obtain the matrix:
 
  [0 0
  0 10
  5 0]
 
  while in R I cannot do that, because x has already been defined as a 2
 by 2 matrix.
  I find all this very inefficient for dealig with matrices.
  This is my point :-)
 
  Francesco
 
 I'm not familiar with R but what you are saying about R in comparison with
 Matlab or Octave is that Matlab/Octave do not check array bounds. Thus, you
 might erroneously write into a matrix element that doesn't exist and
 Matlab/Octave does not report an error. Array bounds checking is the price
 to pay for having safer code. FWIW, Python and Mathematica are examples of
 languages that are only half-safe with respect to array indexing, meaning
 if you write past the beginning of an array (usually negative indexes) it
 assumes you want to index the array backwards, from the end. These are all
 conveniences but of course can also cause erroneous results or increased
 time spent debugging. I personally am happy to write in a safe language
 (Ada) and spend a little bit more time writing code (it's easy to declare a
 new array and copy the old one into it; you can write a subroutine if you
 want) and virtually zero time chasing bugs, but others prefer otherwise and
 that's fine.

 Jerry

Re: LyX+Scilab

[Jerry]

On Jan 25, 2015, at 2:56 PM, Benedict Holland benedict.m.holl...@gmail.com 
wrote:

 On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin 
 francesco.menon...@unibs.it wrote:
 Hello Benedict,
 thank you for your answer.
 I give you an example about how R is difficult for dealing with matrices: if 
 in Matlab (or Scilab or Octave or simila) I write:
 
 x(2,2)=10
 
 the following matrix is automatically created:
 
 [0 0
 0 2]
 
 while in R
 
 x[2,2]=10
 
 gives an error because I first have to declare to R that x is a matrix and, 
 furthermore, give it the right dimension.
 If, afterward, I want to add an element to the matrix x, in Matlab I simply 
 write:
 
 x(3,1)=5
 
 and I obtain the matrix:
 
 [0 0
 0 10
 5 0]
 
 while in R I cannot do that, because x has already been defined as a 2 by 2 
 matrix.
 I find all this very inefficient for dealig with matrices.
 This is my point :-)
 
 Francesco
 
I'm not familiar with R but what you are saying about R in comparison with 
Matlab or Octave is that Matlab/Octave do not check array bounds. Thus, you 
might erroneously write into a matrix element that doesn't exist and 
Matlab/Octave does not report an error. Array bounds checking is the price to 
pay for having safer code. FWIW, Python and Mathematica are examples of 
languages that are only half-safe with respect to array indexing, meaning if 
you write past the beginning of an array (usually negative indexes) it assumes 
you want to index the array backwards, from the end. These are all conveniences 
but of course can also cause erroneous results or increased time spent 
debugging. I personally am happy to write in a safe language (Ada) and spend a 
little bit more time writing code (it's easy to declare a new array and copy 
the old one into it; you can write a subroutine if you want) and virtually zero 
time chasing bugs, but others prefer otherwise and that's fine.

Jerry 

Re: LyX+Scilab

[Benedict Holland]

Naw, the issue is just that R is messy. Actually, for this particular case,
I was chuckling a bit since it isn't wrong how R did it, it is just the way
R does it. C++ is exactly the same. Unfortunately, the answer that I will
give for this thread is that the OP should really just learn R or figure
out a workaround. I have done a huge amount of work in many languages and
basically I would not wish to merge them. Loosely coupled code is much
better than tightly coupled. Reading in a CSV or a PDF is much better style
than generating it and working with it inside of a document. Eventually you
start asking yourself questions like, if I gave this Lyx document to a
friend, would they have any idea what I did and could they figure it out in
under 2 minutes? If the answer is no, I might suggest finding a different
way to attack the problem.

I was thinking about this though. For plots, R is one of the best tools you
can use. Believe you me, illustrator is probably the best but R does a
really good job and is not 800$. What I really recommend for plots is that
you save the plot and load it into the document. I have done this many
times. That way, you are regenerating the same plot every time you compile
your document. Also, as long as you can generate a PDF image, you can
include it.

Now I bet there is a case where the plot should be dynamic with every
compile and if that is the case, I don't really know where to go apart from
learning R... if only for the plotting which is some of the best you will
find anywhere. I actually find the whole idea of integrating R into Lyx to
be interesting but misses the point a bit. For tables I could see it being
very useful but within the boundaries of whatever automatically sets up the
table formatting. For images, I would save it and import it as a pdf. BTW,
I have had no end of problems working with sweave and the stata latex
output. It essentially requires me to rewrite the tables anyway. Problems
range anywhere from writing off the end of the page to generating tables
which are unreadable.

BTW, totally off topic but Python is very versatile. It actually makes much
more sense with the power behind dictionaries, document parsing, document
merging, string manipulations, data management, etc. for it to be included
in Lyx. That I am sure was a feat of engineering.

I honestly don't know what to tell the OP anymore as latex includes many
many ways to integrate PDF images within a document. Generating the PDF
outside of the document and importing it seems like far less work than
trying to get it all working in Lyx.

~Ben

On Wed, Jan 28, 2015 at 10:56 PM, Jerry lancebo...@qwest.net wrote:


 On Jan 25, 2015, at 2:56 PM, Benedict Holland 
 benedict.m.holl...@gmail.com wrote:

  On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin 
 francesco.menon...@unibs.it wrote:
  Hello Benedict,
  thank you for your answer.
  I give you an example about how R is difficult for dealing with
 matrices: if in Matlab (or Scilab or Octave or simila) I write:
 
  x(2,2)=10
 
  the following matrix is automatically created:
 
  [0 0
  0 2]
 
  while in R
 
  x[2,2]=10
 
  gives an error because I first have to declare to R that x is a matrix
 and, furthermore, give it the right dimension.
  If, afterward, I want to add an element to the matrix x, in Matlab I
 simply write:
 
  x(3,1)=5
 
  and I obtain the matrix:
 
  [0 0
  0 10
  5 0]
 
  while in R I cannot do that, because x has already been defined as a 2
 by 2 matrix.
  I find all this very inefficient for dealig with matrices.
  This is my point :-)
 
  Francesco
 
 I'm not familiar with R but what you are saying about R in comparison with
 Matlab or Octave is that Matlab/Octave do not check array bounds. Thus, you
 might erroneously write into a matrix element that doesn't exist and
 Matlab/Octave does not report an error. Array bounds checking is the price
 to pay for having safer code. FWIW, Python and Mathematica are examples of
 languages that are only half-safe with respect to array indexing, meaning
 if you write past the beginning of an array (usually negative indexes) it
 assumes you want to index the array backwards, from the end. These are all
 conveniences but of course can also cause erroneous results or increased
 time spent debugging. I personally am happy to write in a safe language
 (Ada) and spend a little bit more time writing code (it's easy to declare a
 new array and copy the old one into it; you can write a subroutine if you
 want) and virtually zero time chasing bugs, but others prefer otherwise and
 that's fine.

 Jerry

Re: LyX+Scilab

[Jerry]

On Jan 25, 2015, at 2:56 PM, Benedict Holland  
wrote:

> On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin 
>  wrote:
> Hello Benedict,
> thank you for your answer.
> I give you an example about how R is difficult for dealing with matrices: if 
> in Matlab (or Scilab or Octave or simila) I write:
> 
> x(2,2)=10
> 
> the following matrix is automatically created:
> 
> [0 0
> 0 2]
> 
> while in R
> 
> x[2,2]=10
> 
> gives an error because I first have to declare to R that x is a matrix and, 
> furthermore, give it the right dimension.
> If, afterward, I want to add an element to the matrix x, in Matlab I simply 
> write:
> 
> x(3,1)=5
> 
> and I obtain the matrix:
> 
> [0 0
> 0 10
> 5 0]
> 
> while in R I cannot do that, because x has already been defined as a 2 by 2 
> matrix.
> I find all this very inefficient for dealig with matrices.
> This is my point :-)
> 
> Francesco
> 
I'm not familiar with R but what you are saying about R in comparison with 
Matlab or Octave is that Matlab/Octave do not check array bounds. Thus, you 
might erroneously write into a matrix element that doesn't exist and 
Matlab/Octave does not report an error. Array bounds checking is the price to 
pay for having safer code. FWIW, Python and Mathematica are examples of 
languages that are only "half-safe" with respect to array indexing, meaning if 
you write past the beginning of an array (usually negative indexes) it assumes 
you want to index the array backwards, from the end. These are all conveniences 
but of course can also cause erroneous results or increased time spent 
debugging. I personally am happy to write in a safe language (Ada) and spend a 
little bit more time writing code (it's easy to declare a new array and copy 
the old one into it; you can write a subroutine if you want) and virtually zero 
time chasing bugs, but others prefer otherwise and that's fine.

Jerry 

Re: LyX+Scilab

[Benedict Holland]

Naw, the issue is just that R is messy. Actually, for this particular case,
I was chuckling a bit since it isn't wrong how R did it, it is just the way
R does it. C++ is exactly the same. Unfortunately, the answer that I will
give for this thread is that the OP should really just learn R or figure
out a workaround. I have done a huge amount of work in many languages and
basically I would not wish to merge them. Loosely coupled code is much
better than tightly coupled. Reading in a CSV or a PDF is much better style
than generating it and working with it inside of a document. Eventually you
start asking yourself questions like, if I gave this Lyx document to a
friend, would they have any idea what I did and could they figure it out in
under 2 minutes? If the answer is no, I might suggest finding a different
way to attack the problem.

I was thinking about this though. For plots, R is one of the best tools you
can use. Believe you me, illustrator is probably the best but R does a
really good job and is not 800$. What I really recommend for plots is that
you save the plot and load it into the document. I have done this many
times. That way, you are regenerating the same plot every time you compile
your document. Also, as long as you can generate a PDF image, you can
include it.

Now I bet there is a case where the plot should be dynamic with every
compile and if that is the case, I don't really know where to go apart from
learning R... if only for the plotting which is some of the best you will
find anywhere. I actually find the whole idea of integrating R into Lyx to
be interesting but misses the point a bit. For tables I could see it being
very useful but within the boundaries of whatever automatically sets up the
table formatting. For images, I would save it and import it as a pdf. BTW,
I have had no end of problems working with sweave and the stata latex
output. It essentially requires me to rewrite the tables anyway. Problems
range anywhere from writing off the end of the page to generating tables
which are unreadable.

BTW, totally off topic but Python is very versatile. It actually makes much
more sense with the power behind dictionaries, document parsing, document
merging, string manipulations, data management, etc. for it to be included
in Lyx. That I am sure was a feat of engineering.

I honestly don't know what to tell the OP anymore as latex includes many
many ways to integrate PDF images within a document. Generating the PDF
outside of the document and importing it seems like far less work than
trying to get it all working in Lyx.

~Ben

On Wed, Jan 28, 2015 at 10:56 PM, Jerry  wrote:

>
> On Jan 25, 2015, at 2:56 PM, Benedict Holland <
> benedict.m.holl...@gmail.com> wrote:
>
> > On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin <
> francesco.menon...@unibs.it> wrote:
> > Hello Benedict,
> > thank you for your answer.
> > I give you an example about how R is difficult for dealing with
> matrices: if in Matlab (or Scilab or Octave or simila) I write:
> >
> > x(2,2)=10
> >
> > the following matrix is automatically created:
> >
> > [0 0
> > 0 2]
> >
> > while in R
> >
> > x[2,2]=10
> >
> > gives an error because I first have to declare to R that x is a matrix
> and, furthermore, give it the right dimension.
> > If, afterward, I want to add an element to the matrix x, in Matlab I
> simply write:
> >
> > x(3,1)=5
> >
> > and I obtain the matrix:
> >
> > [0 0
> > 0 10
> > 5 0]
> >
> > while in R I cannot do that, because x has already been defined as a 2
> by 2 matrix.
> > I find all this very inefficient for dealig with matrices.
> > This is my point :-)
> >
> > Francesco
> >
> I'm not familiar with R but what you are saying about R in comparison with
> Matlab or Octave is that Matlab/Octave do not check array bounds. Thus, you
> might erroneously write into a matrix element that doesn't exist and
> Matlab/Octave does not report an error. Array bounds checking is the price
> to pay for having safer code. FWIW, Python and Mathematica are examples of
> languages that are only "half-safe" with respect to array indexing, meaning
> if you write past the beginning of an array (usually negative indexes) it
> assumes you want to index the array backwards, from the end. These are all
> conveniences but of course can also cause erroneous results or increased
> time spent debugging. I personally am happy to write in a safe language
> (Ada) and spend a little bit more time writing code (it's easy to declare a
> new array and copy the old one into it; you can write a subroutine if you
> want) and virtually zero time chasing bugs, but others prefer otherwise and
> that's fine.
>
> Jerry

LyX+Scilab

[Francesco Menoncin]

Dear LyXers,
I use and I like very much the package Knitr for executing R commands in 
LyX. Nevertheless, R is not so easy to use for big matrix computation. I 
wonder whether there exists a project similar to Knitr but allowing to 
code in LyX in other languages like Scilab, or Octave, or Matlab.


Thank you

Francesco

--
---
Francesco Menoncin
Associate Professor
University of Brescia
Department of Economics and Management
E-mail: francesco.menon...@unibs.it
Tel. 0039-0302988806
Fax. 0039-0302988836
https://sites.google.com/site/francescomenoncin/
http://www.eco.unibs.it/~menoncin/


--
Informativa sulla Privacy: http://www.unibs.it/node/8155

Re: LyX+Scilab

[Benedict Holland]

R is not so easy to use for big matrix computations? This is a first for
me. R is extremely efficient at dealing with matrices. If you have huge
matrices, I don't think it would perform much better than anything else. If
you need sparse matrix functionality, R comes with it built in. Python is
actually horrible for mathematical computations since it basically runs
everything in FORTRAN and C. The overhead to manipulate the data to call
these libraries is enormous. R might actually be doing the same. Matlab
does this as well.

Honestly, what I have found is that printing tables and results is best
done by manually coding latex using booktabs and three part table. It isn't
that hard to do either. Write your results as a CSV and if you are good at
scripting, just separate the csv from the data frame into rows. Add a
header and you are good to go. This way, any time you have a CSV, you can
easily get it into tabular format and all you have to do is specify the
header and footer, something that is notoriously difficult to automate.
Trust me, I have tried a dozen or more times and it never just works.

~Ben

On Sun, Jan 25, 2015 at 3:35 PM, Steve Burnham dan...@gmail.com wrote:

 There is a package on CTAN called pythontex that allows you to run python
 from within a document.  I know LyX heavily relies on Python as well and I
 think I have seen mention of people using python code in their documents.
 I used to use Matlab for data processing but have completely switched over
 to python so I’m pretty sure you could do everything you want to do in
 python.


 -Steve

 On January 25, 2015 at 7:00:05 AM, Francesco Menoncin (
 francesco.menon...@unibs.it) wrote:

 Dear LyXers,
 I use and I like very much the package Knitr for executing R commands in
 LyX. Nevertheless, R is not so easy to use for big matrix computation. I
 wonder whether there exists a project similar to Knitr but allowing to
 code in LyX in other languages like Scilab, or Octave, or Matlab.

 Thank you

 Francesco

 --
 ---
 Francesco Menoncin
 Associate Professor
 University of Brescia
 Department of Economics and Management
 E-mail: francesco.menon...@unibs.it
 Tel. 0039-0302988806
 Fax. 0039-0302988836
 https://sites.google.com/site/francescomenoncin/
 http://www.eco.unibs.it/~menoncin/


 --
 Informativa sulla Privacy: http://www.unibs.it/node/8155

Re: LyX+Scilab

[Benedict Holland]

So then the problem isn't with inefficient matrix computations but with the
language itself. On that I can wholehearted agree. I would say though, IMHO
Matlab isn't much better, and in fact can be much much worse. I tend to try
to avoid R but it is a very powerful language once you get the basics. What
you might consider doing is dealing with data frames and then converting
into matrices. R is essentially a data frame language with some stuff
stapled on the side, like python is a dictionary language with some stuff
stapled on the side. Matlab is strange with most variables being fully
scoped and that drives me crazy. Honestly, in the long run you will get
many benefits with R that you won't get with Matlab. I recommend picking it
up. Also, I wouldn't expect any proprietary software to be compatible with
lyx. That sort of development is very difficult.

What I would recommend doing is using whatever language you want and saving
whatever it is as a CSV. From there, I would write a script which will
automatically spit out the middle part of a table. Something along the
lines of

Header 1  1  2  3 \\
 (1)  (2)  (3) \\
Header 2  1  2  3 \\
 (1)  (2)  (3) \\

etc. Then all that is required for you to do is start the tabular
environment:

\begin{tabular}{lm{.25cm}...}
Your output here
\end{tabular}

The hardest part that lyx and most automated table generators have
(including and especially R and Stata) is the header which defines how your
table should look and feel. If your table is long, it should use longtable.
If you want table notes, you need to use three part table. Do you want it
to look nice and want full control over line widths and line placement? Use
booktabs. These are highly context dependent and really up to the end user,
making them very difficult to automate leaving someone like you and me with
the requirement to learn how to typeset tables manually. Look at booktabs
and threeparttable for guides. Making a good table takes a very long time
and requires you to become good with typesetting. Choosing to read in a CSV
and export a latex table middle will allow you to use quite literally
whatever software you want, as long as it can produce a CSV. Actually, if
it would help, I might be able to whip up a script sometime soon in python.
It will not be the most robust script in the whole of scripting, but should
at least be usable. I don't really deal with lyx/python interactions as I
prefer to leave the two separate. Perhaps though, with the script, it might
not be so bad and something worth considering.

~Ben

On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin 
francesco.menon...@unibs.it wrote:

  Hello Benedict,
 thank you for your answer.
 I give you an example about how R is difficult for dealing with matrices:
 if in Matlab (or Scilab or Octave or simila) I write:

 x(2,2)=10

 the following matrix is automatically created:

 [0 0
 0 2]

 while in R

 x[2,2]=10

 gives an error because I first have to declare to R that x is a matrix
 and, furthermore, give it the right dimension.
 If, afterward, I want to add an element to the matrix x, in Matlab I
 simply write:

 x(3,1)=5

 and I obtain the matrix:

 [0 0
 0 10
 5 0]

 while in R I cannot do that, because x has already been defined as a 2 by
 2 matrix.
 I find all this very inefficient for dealig with matrices.
 This is my point :-)

 Francesco

 Il 25/01/2015 22:25, Benedict Holland ha scritto:

 R is not so easy to use for big matrix computations? This is a first for
 me. R is extremely efficient at dealing with matrices. If you have huge
 matrices, I don't think it would perform much better than anything else. If
 you need sparse matrix functionality, R comes with it built in. Python is
 actually horrible for mathematical computations since it basically runs
 everything in FORTRAN and C. The overhead to manipulate the data to call
 these libraries is enormous. R might actually be doing the same. Matlab
 does this as well.

  Honestly, what I have found is that printing tables and results is best
 done by manually coding latex using booktabs and three part table. It isn't
 that hard to do either. Write your results as a CSV and if you are good at
 scripting, just separate the csv from the data frame into rows. Add a
 header and you are good to go. This way, any time you have a CSV, you can
 easily get it into tabular format and all you have to do is specify the
 header and footer, something that is notoriously difficult to automate.
 Trust me, I have tried a dozen or more times and it never just works.

  ~Ben

 On Sun, Jan 25, 2015 at 3:35 PM, Steve Burnham dan...@gmail.com wrote:

  There is a package on CTAN called pythontex that allows you to run
 python from within a document.  I know LyX heavily relies on Python as well
 and I think I have seen mention of people using python code in their
 documents.  I used to use Matlab for data processing but have completely
 switched over to python so I’m pretty sure 

Re: LyX+Scilab

[Steve Burnham]

There is a package on CTAN called pythontex that allows you to run python from 
within a document.  I know LyX heavily relies on Python as well and I think I 
have seen mention of people using python code in their documents.  I used to 
use Matlab for data processing but have completely switched over to python so 
I’m pretty sure you could do everything you want to do in python.


-Steve

On January 25, 2015 at 7:00:05 AM, Francesco Menoncin 
(francesco.menon...@unibs.it) wrote:

Dear LyXers,  
I use and I like very much the package Knitr for executing R commands in  
LyX. Nevertheless, R is not so easy to use for big matrix computation. I  
wonder whether there exists a project similar to Knitr but allowing to  
code in LyX in other languages like Scilab, or Octave, or Matlab.  

Thank you  

Francesco  

--  
---  
Francesco Menoncin  
Associate Professor  
University of Brescia  
Department of Economics and Management  
E-mail: francesco.menon...@unibs.it  
Tel. 0039-0302988806  
Fax. 0039-0302988836  
https://sites.google.com/site/francescomenoncin/  
http://www.eco.unibs.it/~menoncin/  


--  
Informativa sulla Privacy: http://www.unibs.it/node/8155  

LyX+Scilab

[Francesco Menoncin]

Dear LyXers,
I use and I like very much the package Knitr for executing R commands in 
LyX. Nevertheless, R is not so easy to use for big matrix computation. I 
wonder whether there exists a project similar to Knitr but allowing to 
code in LyX in other languages like Scilab, or Octave, or Matlab.


Thank you

Francesco

--
---
Francesco Menoncin
Associate Professor
University of Brescia
Department of Economics and Management
E-mail: francesco.menon...@unibs.it
Tel. 0039-0302988806
Fax. 0039-0302988836
https://sites.google.com/site/francescomenoncin/
http://www.eco.unibs.it/~menoncin/


--
Informativa sulla Privacy: http://www.unibs.it/node/8155

Re: LyX+Scilab

[Steve Burnham]

There is a package on CTAN called pythontex that allows you to run python from 
within a document.  I know LyX heavily relies on Python as well and I think I 
have seen mention of people using python code in their documents.  I used to 
use Matlab for data processing but have completely switched over to python so 
I’m pretty sure you could do everything you want to do in python.


-Steve

On January 25, 2015 at 7:00:05 AM, Francesco Menoncin 
(francesco.menon...@unibs.it) wrote:

Dear LyXers,  
I use and I like very much the package Knitr for executing R commands in  
LyX. Nevertheless, R is not so easy to use for big matrix computation. I  
wonder whether there exists a project similar to Knitr but allowing to  
code in LyX in other languages like Scilab, or Octave, or Matlab.  

Thank you  

Francesco  

--  
---  
Francesco Menoncin  
Associate Professor  
University of Brescia  
Department of Economics and Management  
E-mail: francesco.menon...@unibs.it  
Tel. 0039-0302988806  
Fax. 0039-0302988836  
https://sites.google.com/site/francescomenoncin/  
http://www.eco.unibs.it/~menoncin/  


--  
Informativa sulla Privacy: http://www.unibs.it/node/8155  

Re: LyX+Scilab

[Benedict Holland]

R is not so easy to use for big matrix computations? This is a first for
me. R is extremely efficient at dealing with matrices. If you have huge
matrices, I don't think it would perform much better than anything else. If
you need sparse matrix functionality, R comes with it built in. Python is
actually horrible for mathematical computations since it basically runs
everything in FORTRAN and C. The overhead to manipulate the data to call
these libraries is enormous. R might actually be doing the same. Matlab
does this as well.

Honestly, what I have found is that printing tables and results is best
done by manually coding latex using booktabs and three part table. It isn't
that hard to do either. Write your results as a CSV and if you are good at
scripting, just separate the csv from the data frame into rows. Add a
header and you are good to go. This way, any time you have a CSV, you can
easily get it into tabular format and all you have to do is specify the
header and footer, something that is notoriously difficult to automate.
Trust me, I have tried a dozen or more times and it never just works.

~Ben

On Sun, Jan 25, 2015 at 3:35 PM, Steve Burnham dan...@gmail.com wrote:

 There is a package on CTAN called pythontex that allows you to run python
 from within a document.  I know LyX heavily relies on Python as well and I
 think I have seen mention of people using python code in their documents.
 I used to use Matlab for data processing but have completely switched over
 to python so I’m pretty sure you could do everything you want to do in
 python.


 -Steve

 On January 25, 2015 at 7:00:05 AM, Francesco Menoncin (
 francesco.menon...@unibs.it) wrote:

 Dear LyXers,
 I use and I like very much the package Knitr for executing R commands in
 LyX. Nevertheless, R is not so easy to use for big matrix computation. I
 wonder whether there exists a project similar to Knitr but allowing to
 code in LyX in other languages like Scilab, or Octave, or Matlab.

 Thank you

 Francesco

 --
 ---
 Francesco Menoncin
 Associate Professor
 University of Brescia
 Department of Economics and Management
 E-mail: francesco.menon...@unibs.it
 Tel. 0039-0302988806
 Fax. 0039-0302988836
 https://sites.google.com/site/francescomenoncin/
 http://www.eco.unibs.it/~menoncin/


 --
 Informativa sulla Privacy: http://www.unibs.it/node/8155

Re: LyX+Scilab

[Benedict Holland]

So then the problem isn't with inefficient matrix computations but with the
language itself. On that I can wholehearted agree. I would say though, IMHO
Matlab isn't much better, and in fact can be much much worse. I tend to try
to avoid R but it is a very powerful language once you get the basics. What
you might consider doing is dealing with data frames and then converting
into matrices. R is essentially a data frame language with some stuff
stapled on the side, like python is a dictionary language with some stuff
stapled on the side. Matlab is strange with most variables being fully
scoped and that drives me crazy. Honestly, in the long run you will get
many benefits with R that you won't get with Matlab. I recommend picking it
up. Also, I wouldn't expect any proprietary software to be compatible with
lyx. That sort of development is very difficult.

What I would recommend doing is using whatever language you want and saving
whatever it is as a CSV. From there, I would write a script which will
automatically spit out the middle part of a table. Something along the
lines of

Header 1  1  2  3 \\
 (1)  (2)  (3) \\
Header 2  1  2  3 \\
 (1)  (2)  (3) \\

etc. Then all that is required for you to do is start the tabular
environment:

\begin{tabular}{lm{.25cm}...}
Your output here
\end{tabular}

The hardest part that lyx and most automated table generators have
(including and especially R and Stata) is the header which defines how your
table should look and feel. If your table is long, it should use longtable.
If you want table notes, you need to use three part table. Do you want it
to look nice and want full control over line widths and line placement? Use
booktabs. These are highly context dependent and really up to the end user,
making them very difficult to automate leaving someone like you and me with
the requirement to learn how to typeset tables manually. Look at booktabs
and threeparttable for guides. Making a good table takes a very long time
and requires you to become good with typesetting. Choosing to read in a CSV
and export a latex table middle will allow you to use quite literally
whatever software you want, as long as it can produce a CSV. Actually, if
it would help, I might be able to whip up a script sometime soon in python.
It will not be the most robust script in the whole of scripting, but should
at least be usable. I don't really deal with lyx/python interactions as I
prefer to leave the two separate. Perhaps though, with the script, it might
not be so bad and something worth considering.

~Ben

On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin 
francesco.menon...@unibs.it wrote:

  Hello Benedict,
 thank you for your answer.
 I give you an example about how R is difficult for dealing with matrices:
 if in Matlab (or Scilab or Octave or simila) I write:

 x(2,2)=10

 the following matrix is automatically created:

 [0 0
 0 2]

 while in R

 x[2,2]=10

 gives an error because I first have to declare to R that x is a matrix
 and, furthermore, give it the right dimension.
 If, afterward, I want to add an element to the matrix x, in Matlab I
 simply write:

 x(3,1)=5

 and I obtain the matrix:

 [0 0
 0 10
 5 0]

 while in R I cannot do that, because x has already been defined as a 2 by
 2 matrix.
 I find all this very inefficient for dealig with matrices.
 This is my point :-)

 Francesco

 Il 25/01/2015 22:25, Benedict Holland ha scritto:

 R is not so easy to use for big matrix computations? This is a first for
 me. R is extremely efficient at dealing with matrices. If you have huge
 matrices, I don't think it would perform much better than anything else. If
 you need sparse matrix functionality, R comes with it built in. Python is
 actually horrible for mathematical computations since it basically runs
 everything in FORTRAN and C. The overhead to manipulate the data to call
 these libraries is enormous. R might actually be doing the same. Matlab
 does this as well.

  Honestly, what I have found is that printing tables and results is best
 done by manually coding latex using booktabs and three part table. It isn't
 that hard to do either. Write your results as a CSV and if you are good at
 scripting, just separate the csv from the data frame into rows. Add a
 header and you are good to go. This way, any time you have a CSV, you can
 easily get it into tabular format and all you have to do is specify the
 header and footer, something that is notoriously difficult to automate.
 Trust me, I have tried a dozen or more times and it never just works.

  ~Ben

 On Sun, Jan 25, 2015 at 3:35 PM, Steve Burnham dan...@gmail.com wrote:

  There is a package on CTAN called pythontex that allows you to run
 python from within a document.  I know LyX heavily relies on Python as well
 and I think I have seen mention of people using python code in their
 documents.  I used to use Matlab for data processing but have completely
 switched over to python so I’m pretty sure 

LyX+Scilab

[Francesco Menoncin]

Dear LyXers,
I use and I like very much the package Knitr for executing R commands in 
LyX. Nevertheless, R is not so easy to use for big matrix computation. I 
wonder whether there exists a project similar to Knitr but allowing to 
code in LyX in other languages like Scilab, or Octave, or Matlab.


Thank you

Francesco

--
---
Francesco Menoncin
Associate Professor
University of Brescia
Department of Economics and Management
E-mail: francesco.menon...@unibs.it
Tel. 0039-0302988806
Fax. 0039-0302988836
https://sites.google.com/site/francescomenoncin/
http://www.eco.unibs.it/~menoncin/


--
Informativa sulla Privacy: http://www.unibs.it/node/8155

Re: LyX+Scilab

[Steve Burnham]

There is a package on CTAN called pythontex that allows you to run python from 
within a document.  I know LyX heavily relies on Python as well and I think I 
have seen mention of people using python code in their documents.  I used to 
use Matlab for data processing but have completely switched over to python so 
I’m pretty sure you could do everything you want to do in python.


-Steve

On January 25, 2015 at 7:00:05 AM, Francesco Menoncin 
(francesco.menon...@unibs.it) wrote:

Dear LyXers,  
I use and I like very much the package Knitr for executing R commands in  
LyX. Nevertheless, R is not so easy to use for big matrix computation. I  
wonder whether there exists a project similar to Knitr but allowing to  
code in LyX in other languages like Scilab, or Octave, or Matlab.  

Thank you  

Francesco  

--  
---  
Francesco Menoncin  
Associate Professor  
University of Brescia  
Department of Economics and Management  
E-mail: francesco.menon...@unibs.it  
Tel. 0039-0302988806  
Fax. 0039-0302988836  
https://sites.google.com/site/francescomenoncin/  
http://www.eco.unibs.it/~menoncin/  


--  
Informativa sulla Privacy: http://www.unibs.it/node/8155  

Re: LyX+Scilab

[Benedict Holland]

R is not so easy to use for big matrix computations? This is a first for
me. R is extremely efficient at dealing with matrices. If you have huge
matrices, I don't think it would perform much better than anything else. If
you need sparse matrix functionality, R comes with it built in. Python is
actually horrible for mathematical computations since it basically runs
everything in FORTRAN and C. The overhead to manipulate the data to call
these libraries is enormous. R might actually be doing the same. Matlab
does this as well.

Honestly, what I have found is that printing tables and results is best
done by manually coding latex using booktabs and three part table. It isn't
that hard to do either. Write your results as a CSV and if you are good at
scripting, just separate the csv from the data frame into rows. Add a
header and you are good to go. This way, any time you have a CSV, you can
easily get it into tabular format and all you have to do is specify the
header and footer, something that is notoriously difficult to automate.
Trust me, I have tried a dozen or more times and it never just works.

~Ben

On Sun, Jan 25, 2015 at 3:35 PM, Steve Burnham  wrote:

> There is a package on CTAN called pythontex that allows you to run python
> from within a document.  I know LyX heavily relies on Python as well and I
> think I have seen mention of people using python code in their documents.
> I used to use Matlab for data processing but have completely switched over
> to python so I’m pretty sure you could do everything you want to do in
> python.
>
>
> -Steve
>
> On January 25, 2015 at 7:00:05 AM, Francesco Menoncin (
> francesco.menon...@unibs.it) wrote:
>
> Dear LyXers,
> I use and I like very much the package Knitr for executing R commands in
> LyX. Nevertheless, R is not so easy to use for big matrix computation. I
> wonder whether there exists a project similar to Knitr but allowing to
> code in LyX in other languages like Scilab, or Octave, or Matlab.
>
> Thank you
>
> Francesco
>
> --
> ---
> Francesco Menoncin
> Associate Professor
> University of Brescia
> Department of Economics and Management
> E-mail: francesco.menon...@unibs.it
> Tel. 0039-0302988806
> Fax. 0039-0302988836
> https://sites.google.com/site/francescomenoncin/
> http://www.eco.unibs.it/~menoncin/
>
>
> --
> Informativa sulla Privacy: http://www.unibs.it/node/8155
>
>

Re: LyX+Scilab

[Benedict Holland]

So then the problem isn't with inefficient matrix computations but with the
language itself. On that I can wholehearted agree. I would say though, IMHO
Matlab isn't much better, and in fact can be much much worse. I tend to try
to avoid R but it is a very powerful language once you get the basics. What
you might consider doing is dealing with data frames and then converting
into matrices. R is essentially a data frame language with some stuff
stapled on the side, like python is a dictionary language with some stuff
stapled on the side. Matlab is strange with most variables being fully
scoped and that drives me crazy. Honestly, in the long run you will get
many benefits with R that you won't get with Matlab. I recommend picking it
up. Also, I wouldn't expect any proprietary software to be compatible with
lyx. That sort of development is very difficult.

What I would recommend doing is using whatever language you want and saving
whatever it is as a CSV. From there, I would write a script which will
automatically spit out the middle part of a table. Something along the
lines of

Header 1 & 1 & 2 & 3 \\
& (1) & (2) & (3) \\
Header 2 & 1 & 2 & 3 \\
& (1) & (2) & (3) \\

etc. Then all that is required for you to do is start the tabular
environment:

\begin{tabular}{l>m{.25cm}...}

\end{tabular}

The hardest part that lyx and most automated table generators have
(including and especially R and Stata) is the header which defines how your
table should look and feel. If your table is long, it should use longtable.
If you want table notes, you need to use three part table. Do you want it
to look nice and want full control over line widths and line placement? Use
booktabs. These are highly context dependent and really up to the end user,
making them very difficult to automate leaving someone like you and me with
the requirement to learn how to typeset tables manually. Look at booktabs
and threeparttable for guides. Making a good table takes a very long time
and requires you to become good with typesetting. Choosing to read in a CSV
and export a latex table middle will allow you to use quite literally
whatever software you want, as long as it can produce a CSV. Actually, if
it would help, I might be able to whip up a script sometime soon in python.
It will not be the most robust script in the whole of scripting, but should
at least be usable. I don't really deal with lyx/python interactions as I
prefer to leave the two separate. Perhaps though, with the script, it might
not be so bad and something worth considering.

~Ben

On Sun, Jan 25, 2015 at 4:38 PM, Francesco Menoncin <
francesco.menon...@unibs.it> wrote:

>  Hello Benedict,
> thank you for your answer.
> I give you an example about how R is difficult for dealing with matrices:
> if in Matlab (or Scilab or Octave or simila) I write:
>
> x(2,2)=10
>
> the following matrix is automatically created:
>
> [0 0
> 0 2]
>
> while in R
>
> x[2,2]=10
>
> gives an error because I first have to declare to R that x is a matrix
> and, furthermore, give it the right dimension.
> If, afterward, I want to add an element to the matrix x, in Matlab I
> simply write:
>
> x(3,1)=5
>
> and I obtain the matrix:
>
> [0 0
> 0 10
> 5 0]
>
> while in R I cannot do that, because x has already been defined as a 2 by
> 2 matrix.
> I find all this very inefficient for dealig with matrices.
> This is my point :-)
>
> Francesco
>
> Il 25/01/2015 22:25, Benedict Holland ha scritto:
>
> R is not so easy to use for big matrix computations? This is a first for
> me. R is extremely efficient at dealing with matrices. If you have huge
> matrices, I don't think it would perform much better than anything else. If
> you need sparse matrix functionality, R comes with it built in. Python is
> actually horrible for mathematical computations since it basically runs
> everything in FORTRAN and C. The overhead to manipulate the data to call
> these libraries is enormous. R might actually be doing the same. Matlab
> does this as well.
>
>  Honestly, what I have found is that printing tables and results is best
> done by manually coding latex using booktabs and three part table. It isn't
> that hard to do either. Write your results as a CSV and if you are good at
> scripting, just separate the csv from the data frame into rows. Add a
> header and you are good to go. This way, any time you have a CSV, you can
> easily get it into tabular format and all you have to do is specify the
> header and footer, something that is notoriously difficult to automate.
> Trust me, I have tried a dozen or more times and it never just works.
>
>  ~Ben
>
> On Sun, Jan 25, 2015 at 3:35 PM, Steve Burnham  wrote:
>
>>  There is a package on CTAN called pythontex that allows you to run
>> python from within a document.  I know LyX heavily relies on Python as well
>> and I think I have seen mention of people using python code in their
>> documents.  I used to use Matlab for data