I would like to able to plot the progress of a MIP solved by cplex. Specifically I would like to plot lower and upper bounds as functions of cpu-time. But copying an pasting from the node log does not seem to be a smartest way of proceeding. Is it possible to access these information and to print them out/to file during the optimization?
I am using the concert technology C++ interface.
You can add a "MIP info callback" using the API routine CPXsetinfocallbackfunc or its analogue in Concert.
(Copying-and-pasting the log it dumps to the terminal is perfectly fine for getting a rough idea of what's going on, but be aware that the results can be highly variable.)
Callbacks are what you are looking for. You can find a nice introduction here:
http://eaton.math.rpi.edu/cplex90html/usrcplex/callbacks.html
Related
I have a few complex classes generated by a third party domain-specific tool. The classes are complex because the tool tries to be generic as much as possible, so that I was asked to design a GDB pretty printer python script to improve the debugging experience.
I designed it by "to_string" APIs which can print the information with organized rich text (e.g. colors), so far everything works fine.
However, I don't know how to get the output format user requests, no matter that user uses "p" or "p/x" will all output the same string because my script does not know user wants hex rather than decimal.
I tried googling but didn't figure out an elegant approach. I indeed have a few workarounds but they all change the usages (for example, implement two printers to be switched), please give me some suggestions, thank you.
Just realized there had been a feature request to GDB already.
Bug 17291 - IWBN if the print format was available to pretty-printers.
I am learning how to use quantlib to price derivatives. What is the best way to output some of the Quantlib specific classes to console window? For example
shared_ptr<YieldTermStructure> forwardCurve(new InterpolatedDiscountCurve<LogLinear>(dates,discountFactor,Actual360()));
Handle<YieldTermStructure> forwardingTermStructure(forwardCurve);
shared_ptr<IborIndex> euribor(new Euribor(3*Months,forwardingTermStructure));
What will be the best way for me to output forwardCurve to and euribor to console window? Then I can see the intermediate result of the code to see if they are as expected.
Many thanks.
There's no predefined way to output those classes to console, but you can use their inspector to output the relevant data (so, e.g., you can call forwardCurve->times() and forwardCurve->discounts() to inspect the values you're interpolating) or you can call their methods to see the results of their calculations (e.g., forwardCurve->discount(d) to retrieve the discount factor at a given date, or euribor->fixing(d) to retrieve the expected index fixing). The returned values can be written to console.
As an alternative, you might consider stepping through the code inside a debugger. In modern IDEs, this will give you the same information more easily.
I am trying to create C++ code that allows User Input in selecting a variety of fields, then it will calculate many different angles and show the users the results, as well as a graph.
However, it has been suggested to us by our lecturer that it may be a good idea to write the code to these calculations etc in C++, then input the results into Excel.
Does anyone have any idea how to do this? Literally looking for a way for the user to fill in the required values on C++ and then to be AUTOMATICALLY taken to the excel file to show the results in the table and graph format.
If this is not possible, is there a way to display the results in the table and graph format through C++?
Thanks very much in advance
Excel provides COM interface which you can use from your C++ application.
This can be done in the way described in this article:
http://support.microsoft.com/kb/216686
This link might also be useful:
http://www.codeproject.com/Articles/10886/How-to-use-Managed-C-to-Automate-Excel
I think the second link would be better for you as its more of a step by step guide which should help you to workout the answer.
Use COM Automation to automate excel.
The best way to do this is to use the vole library by Matthew Wilson at
http://vole.sourceforge.net/
Take a look at the examples. I do not think there is an example for excel, but there is one for microsoft word at http://www.codeproject.com/KB/COM/VOLE_word.aspx
I have used vole in the past, and it makes it a whole lot easier
I was trying to find if there exists a library or tool that will allow me to visually debug my program. i.e. something that shows a graphviz like tree structure and highlights exactly where I am in the process tree at a breakpoint. This would give a faster understanding of how my process works rather than sequentially debug through and create a tree in my mind.
I found something that partially does what I am looking for, i.e. show a tree structure of my process and the number of calls made per function call
http://www.ibm.com/developerworks/library/l-graphvis/
If it doesn't exist then I might plan on writing something that does the job. Thanks
-CV
The debug visualization plugin for Eclipse sounds like something that might be helpful for you. Furthermore, the venerable Data Display Debugger also has some automatic routines for creating graphs, albeit of the data structures you are currently seeing. I also like the visualization of kcachegrind, but it is not exactly a debugging aid. However, its graphical view shows you the position in the execution tree.
Since there does not seem to be a tool that matches your requirements exactly, maybe these ones will inspire you to write your own ;)
I have frequently encounter the following debugging scenario:
Tester provide some reproduce steps for a bug. And to find out where the problem is, I try to play with these reproduce steps to get the minimum necessary reproduce steps. Sometimes, luckily I found that when do a minor change to the steps, the problem is gone.
Then the job turns to find the difference in code workflow between these two reproduce steps. This job is tedious and painful especially when you are working on a large code base and it go through a lot code and involve lots of state changes which you are not familiar with.
So I was wondering is there any tools available to compare "code workflow". As I've learned the "wt" command in WinDbg, I thought it might be possible to do it. For example, I can run the "wt" command on some out most functions with 2 different reproduce steps and then compare the difference between outputs. Then it should be easy to found where the code flow starts to diverge.
But the problem with WinDBG is "wt" is quite slow (maybe I should use a log file instead of output to screen) and not very user-friendly (compared with visual studio debugger) ... So I want to ask you guys is there any existing tools available . or is it possible and difficult to develop a "plug-in" for visual studio debugger to support this functionality ?
Thanks
I'd run it under a profiler in "coverage" mode, then use diff on the results to see which parts of the code were executed in one run by not the other.
Sorry, I don't know of a tool which can do what you want, but even if it existed it doesn't sound like the quickest approach to finding out where the lower layer code is failing.
I would recommend to instrument your layer's code with high-level logs so you can know which module fails, stalls, etc. In debug, your logger can write to file, to output debug window, etc.
In general, failing fast and using exceptions are good ways to find out easily where things go bad.
Doing something after the fact is not going to cut it, since your problem is reproducing it.
The issue with bugs is seldom some interal wackiness but usually what the user's actually doing. If you log all the commands that the user enters then they can simply send you the log. You can substitute button clicks, mouse selects, etc. This will have some cost but certainly much less than something that keeps track of every method visited.
I am assuming that if you have a large application that you have good logging or tracing.
I work on a large server product with over 40 processes and over one million lines of code. Most of the time the error in the trace file is enough to identify the location of problem. However sometimes the error I see in the trace file is caused by some earlier code and the reason for this can be hard to spot. Then I use a comparative debugging technique:
Reproduce the first scenario, copy the trace to a new file (if the application is multi threaded ensure you only have the trace for the thread that does the work).
Reproduce the second scenario, copy the trace to a new file.
Remove the timestamps from the log files (I use awk or sed for this).
Compare the log files with winmerge or similar, to see where and how they diverge.
This technique can be a little time consuming, but is much quicker than stepping through thousand of lines in the debugger.
Another useful technique is producing uml sequence diagrams from trace files. For this you need the function entry and exit positions logged consistently. Then write a small script to parse your trace files and use sequence.jar to produce uml diagrams as png files. This is a great way to understand the logic of code you haven't touched in a while. I wrapped a small awk script in a batch file, I just provide trace file and line number to start then it untangles the threads and generates the input text to sequence.jar then runs its to create the uml diagram.