March 2009 - Boost - lists.boost.org

Re: [boost] [math] Does any one interesting in calculating derivative for a expression automatically?
by Dongfei Yin 06 Mar '09

06 Mar '09

Hi Chris, The basic theory is same, but to do it at compile time, it can not be as simple as run time. CppAD do it during run time. type in run time is not so important as compile time. It can be done not only by operator overloading, but also by template specializing. As I say, when you get d(_x * _x, _x), that is a functor, then you want d(d(_x * _x, _x), _x), you see that there is no chance for operator to do anything, that means the result of d(_x * _x, _x) should be simplified before d(d(_x * _x, _x), _x) by template secializeing. Another example, Define hypersphere<N> as (pi^(N/2)*r^N) / gamma(N/2 + 1), suppose N is known at compile time,if all the type are under control, think about (double_<pi>^(int_<N>/2)*var(r)^int_<N> / gamma(int_<N)>/2 + 1), all the type can be controlled, then N/2 can be calculated during compile time. by if just using operator overloading, it can only be done during run time. you can not overload operator+(int, int) (just a example, actually, it can be optimized by compiler automatically). Dongfei. > While not strictly the same, there are all sorts of C++ based packages > out there for doing this kind of thing. The approach taken is > 'automatic differentiation', and it basically works by applying the > chain rule to every individual step in a sequence of operations. One > approach to doing this is through operator overloading, and replacing > the underlying 'double' class with something that maintains the > appropriate state. There are already some nice open source packages out > there for doing this. More info: > > http://en.wikipedia.org/wiki/Automatic_differentiation > http://www.autodiff.org/ > http://www.coin-or.org/CppAD/ > > Cheers, > > Chris >

1 0

Re: [boost] [math] Does any one interesting in calculating derivative for a expression automatically?
by Dongfei Yin 06 Mar '09

06 Mar '09

Hi Bruno, Till now, it can only work with polynomial and trigonometric functions. and I have a good idea of simplifying a expression. as you said, if N is compile time known, it should be int_<N>, then the expression with it can be simplified. As the hypersphere talked before, if N=2, and the volume expression will be simplified from (int_<N>^(int_<N>/int_<2>)*var(r)^int_<N>) / gamma(int_<N>/int_<2> + int_<1>) to int_<pi>*var(r)^int_<2> if gamma() has been specialized. Till now, (_x*_x*_x) is a functor, it can be simplified to something like pow(_x, int_<3>). without using d(...), it can be used to simplify expression as we talked before. this is another use of it. You know, there are too many functions in math, so I think I can't implement all of them, the functions included in standard library is enough for me. and I will make it easy to be extended. what do you think about it, I mean how many functions should I implement. My main purpose is calculating derivative easily, of course simplify expression is a important part, I am working on it. Dongfei.

1 0

Boost package maintenance for Cygwin
by bnv＠nc.rr.com 06 Mar '09

06 Mar '09

I make use of Boost in a Cygwin environment and was happy to see it supported as a package. Unfortunately, the current package is based on 1.33.1 and there have been many updates since. The maintainer has indicated difficulty working with the Boost distribution (see below, from the main cygwin mailing list), and has stopped supporting the package. I'm not the right person to take it over, but have an interest in seeing it maintained, and thought it worth bringing this to the attention of the boost community to potentially look into the issue(s). Thanks, Brian > I have given up on Boost package on Cygwin. Boost build AFAIK does not > follow the naming conventions Cygwin has for DLLs (or did not the last > time I checked) and I do not have the willpower to hack it into submission. > > If you feel like taking over the Cygwin Boost package, try Boost Cygport > as starting point for a new Cygwin package. > > - -- > wilx

2 1

Re: [boost] [math] Does any one interesting in calculating derivative for a expression automatically?
by Dongfei Yin 06 Mar '09

06 Mar '09

Hi Joel, I it is easy to map it onto d<N>(f,v), but I think it is difficult to be optimized for N=1, 2, 3 ... because to calculate the derivative of d<5>(f,v), actually, d<1>, d<2> ... d<5> will be calculated . but after each derivative is calculated, it will be simplified before next step. now the result of derivative expression can be saved as a functor, so d(d(x*x,x),x), y = d(x*x,x); d(y,x); is one of my purpose. I will implement d<N>(...), that is a good idea and convenience, thank you. Dongfei. > Can't this be mapped onto a d<N>(f,v) function that is optimized for N = > 1,2,3,... ? > I bet that if you use proto, matching d(d(d( ... )))) and repalcing by > d<N>(...) should be easy. > > Do you perform your derivation recursively on the derivative degree (ie > d(d(x*x,x),x) is y = d(x*x,x); d(y,x);) > If yes, you may want to use coarser grain. Let's say you specialized d() > d<2>() and d<3>() for ex. Computing d<5> > is d<2>(d<3>()) instead of d(d(d(d(d()))). Of course, using d,d2, d4 is > maybe a better option so you cna use bit representaiton of derivative > degree to find how to combine the derivative. >

1 0

An analogue of boost.function with multiple signatures
by Ji�� Pale�ek 06 Mar '09

06 Mar '09

Hello, I'd like to have an object similar to boost::function, but with multiple function signatures. Eg. something like that multifunction<std::string(std::string), float(float), unsigned(unsigned)> mf; mf=_1 + _1; // I mean boost::lambda::_1 mf("ab"); // should be "abab" mf((UINT_MAX/2)+1); // should be 0 mf((UINT_MAX/2)+1.0); // should be UINT_MAX+1.0 Would it be possible/feasible to change boost::function to allow this (as opposed to implementing it separately from boost::function)? Was something like this even proposed before (I vaguely recall something, but couldn't google anything useful, so I may be wrong)? Regards Jiri Palecek

2 2

BGL question about districting
by Barend Gehrels 06 Mar '09

06 Mar '09

Hi, We're about an automated districting tool. We want to create Y districts from X polygons, all having a specific value (e.g. area or inhabitants), where the total value of each district is as similar as possible and each district is also as similar as possible. Let's call the value the "potential". Of course this problem is NP-hard so we're glad with the first solution or e.g. the first ten solutions. We thought to implement this by the Boost Graph Library. First we detect neighbour relations with our own Geometry Library (using the still-to-build "touches" relationship). We've a graph then and can divide the graph into subgraphs, each having connected nodes and an (nearly) equal "potential". I've some experience with BGL, having used the "connected_components". But I cannot find an appropriate algorithm for this problem, if it is available. Actually it even could be a "network flow" algorithm dividing the total potential by Y, and starting at some point flowing the potential through the graph. But all network flow algorithms seem to be on directed graphs. This is an undirected graph. It it is not possible with a standard BGL algorithm I think we can still solve it using BGL and e.g. breadth-sort-first, but just want to verify if it is possible or maybe by an unexpected workaround. Thanks, Barend

4 5

Blocking Queue implementation in C++
by Filip Klasson 06 Mar '09

06 Mar '09

Hi, I need to create a Blocking queue implementation in *C++* with the help of Boost libraries and i am looking for some advise. What i need is a queue with linked nodes. The queue needs to be ordered as FIFO (first-in-first-out). I need the following functionality (Functionality from BlockingQueue implementation in Java) INSERT: Insert new items into the queue and if there is no space the queue needs to wait until space is available. RETRIEVE: Retrieve and removes the element that has waited the longest. If there is no elements in the queue it waits until an element becomes available. SIZE: Return the number of elements in the queue. POLL: The same as retrieve but only waits for a specified time for items to become available. I found a messageQueue that looks something like what i need (found here: http://www.boost.org/doc/libs/1_36_0/boost/interprocess/ipc/message_queue.h…) is that correct? /Filip

1 0

[itl] Preview 4 of the Interval Template Library
by Joachim Faulhaber 05 Mar '09

05 Mar '09

Dear list, I am happy to announce that the Interval Template Library (Boost.Itl) converges to a state, that conforms the standards of the boost libraries. It now contains substantial test suites and almost complete quickbook documentation. I hope it will soon be mature for a formal review after a last round of feedback and subsequent improvements. Since the last progress report in November last year the following work has been done. + Interoperability of interval_containers: Addition, subtraction, intersection and symmetric difference are now interoperable for all combinations of interval containers that make sense: You may for instance intersect a split_interval_map with an interval_map, an interval_set an interval or an element. + Added symmetric difference and renamed operators in accordance to other set class templates like e.g. boost::dynamic_bitset. + Introduced infix operators as itl::namespace global functions that are interoperable between interval_maps, sets, intervals and elements, so far as the combinations make sense. + Operators are now +=, +, |=, | : Set union / Map union with aggregation -=, - : Set difference / Map union with aggregation &=, & : Set intersection / Map intersection with aggregation ^=, ^ : Set symmetric difference / Map symmetric difference with aggregation. + Extended the test suite, specifically for the new operators and their overloadings. + Wrote quickbook documentation for the library. Please note that the original ITL as presented in May 2008 has been split into 3 parts and only the *first* part is prepared for boost now: itl : Core library, interval containers, PREPARED FOR BOOST itl_xt : Extensions like histories and generalized crosstables validate: The law based test automaton (LaBatea). A validation tool using laws aka axioms. So please focus on the core part (itl). Download the current version 'preview 4' from the boost vault's containers section: itl.zip http://www.boostpro.com/vault/index.php?action=downloadfile&filename=itl.zi… If you are interested in extended parts or LaBatea you may download all parts as itl_plus.zip http://www.boostpro.com/vault/index.php?action=downloadfile&filename=itl_pl… The quickbook generated documentation is included in the zip files and is also availabe online at http://herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/index.html Sources are also available form the sandbox: https://svn.boost.org/svn/boost/sandbox/itl/ The documentation has a focus on tree aspects (1) A short introduction, that wants to be easy to understand and to provide good and use case related examples: http://herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/index.html#boost… http://herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/boost_itl/exampl… (2) Synoptical representations that allows to see the complete 'picture' of the library. For instance the function matrix http://herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/boost_itl/interf… is supposed to give a complete overview over the most important functions across all itl containers. (3) A description of the semantics which is characterized by sets of laws that apply for the different interval containers. The semantical analysis, based on sets of laws, was a quite interesting work, where software development sometimes transformed into software discovery guided by the validation tool LaBatea ;) My main finding out of that process was, that the semantics of the aggregating itl::Maps is determined by the semantics of their data or codomain parameter U, so that if U is model of concept C then itl::Map<T,U> is model of C which is stated in section concept induction: http://herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/boost_itl/semant… All your feedback is welcome. Cheers Joachim

3 4

[time series] What happened to this library?
by Christian Henning 05 Mar '09

05 Mar '09

Hi there, as far as I can tell this lib has been approved but it doesn't show up in boost. Why is it not in there? Thanks, Christian

11 21

Trunk Inspection Report fixed + other reminders
by Beman Dawes 05 Mar '09

05 Mar '09

The trunk Inspection Report is now cycling daily, with a new URL. See http://boost.cowic.de/rc/inspect-trunk.html The release branch Inspection Report also cycles daily. See http://boost.cowic.de/rc/inspect-release.html The links to these and other useful "State of Boost" links are found at http://www.boost.org/development/testing.html --Beman

2 1