http://creativemachines.cornell.edu/eureqa/
http://code.google.com/p/eureqa-api/
Is there a way of interfacing Eureqa with R?
(windows os)
Perhaps:
Rcpp package and:
The simplest possible complete C++ program using the Eureqa API:
#include eureqa/eureqa.h
#include iostream
int main(int argc, char *argv[])
{
// initialize data set and options
eureqa::data_set data("my_data.txt");
eureqa::search_options options("y = f(x)");
// connect to a eureqa server
eureqa::connection conn("127.0.0.1");
conn.send_data_set(data);
conn.send_options(options);
conn.start_search();
// display results
eureqa::search_progress progress;
eureqa::solution_frontier solutions;
while (conn.query_progress(progress))
{
solutions.add(progress.solution_);
std::cout << progress.summary() << std::endl;
std::cout << solutions.to_string() << std::endl;
}
return 0;
}
Related
Using Poco-1.12.4-release and g++ (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0, The recommended example code for DatagramSocket Receive in this doc:
https://pocoproject.org/slides/200-Network.pdf
compiles with this error:
dgs.cpp: In function ‘int main(int, char**)’:
dgs.cpp:34:33: error: no matching function for call to ‘Poco::Net::DatagramSocket::DatagramSocket(Poco::Net::SocketAddress&)’
34 | Poco::Net::DatagramSocket dgs(sa);
What happened? Is there a fix for this example?
Thanks
The code is:
// DatagramSocket receive example
#include "Poco/Net/DatagramSocket.h"
#include "Poco/Net/SocketAddress.h"
#include <iostream>
int main(int argc, char** argv)
{
Poco::Net::SocketAddress sa(Poco::Net::IPAddress(), 514);
Poco::Net::DatagramSocket dgs(sa);
char buffer[1024];
for (;;)
{
Poco::Net::SocketAddress sender;
int n = dgs.receiveFrom(buffer, sizeof(buffer)-1, sender);
buffer[n] = '\0';
std::cout << sender.toString() << ": " << buffer << std::endl;
}
return 0;
}
I tried this:
Poco::Net::SocketAddress sa(Poco::Net::IPAddress(), 514);
Poco::Net::DatagramSocket dgs(Poco::Net::SocketAddress::IPv4);
dgs.connect(sa);
It compiles, but is it correct?
Code:
using ColumnIndexVector = std::vector<int>;
using ByteVector = std::vector<BYTE>;
void CCreateReportDlg::GetColumnIndexesToExclude()
{
const CString strSection = theApp.GetActiveScheduleSection(_T("Options"));
ByteVector vData = theApp.GetProfileVector(strSection, _T("AssignStatesEx"));
ColumnIndexVector vTemp(vData.begin(), vData.end()); // This converts BYTE to int
m_vColumnIndexesToExclude = vTemp;
}
Is there any way to avoid the requirement for vTemp without manually iterating vData and casting from BYTE to int?
Yes, just use assign(). IDK if you need to use clear() as well, but probably not. Just step through the runtime code the first time to know.
m_vColumnIndexesToExclude.assign(vData.begin(), vData.end());
Here's a test program:
#include <windows.h>
#include <iostream>
#include <vector>
using namespace std;
using ColumnIndexVector = std::vector<int>;
using ByteVector = std::vector<BYTE>;
int main(int argc, char* argv[])
{
cout << "Test" << endl;
ByteVector bytes = {'A', 'B', 'C', 'D'};
ColumnIndexVector colVector;
for ( auto _val: bytes)
{
cout << _val << endl;
}
colVector.assign(bytes.begin(), bytes.end());
for ( auto _val : colVector)
{
cout << _val << endl;
}
return 0;
}
Is it "expected" for QSharedPointer::create() not to work or is it a bug? I get an error:
/usr/include/qt5/QtCore/qsharedpointer_impl.h:439:9: error:
invalid conversion from ‘const void*’ to ‘void*’ [-fpermissive]
new (result.data()) T(std::forward<Args>(arguments)...);
casting from nonconst shared pointer and constructor from raw const pointer work.
I got this for Qt5.7.0 and Qt5.10.0.
Here is a minimal example:
#include <QSharedPointer>
struct A {};
int main(int argc, char *argv[])
{
auto ca = QSharedPointer<const A>::create();
return 0;
}
Here is one file (not minimal) example but with few working cases, 2 not working and a debug. Commented defines are for "not compiling" parts.
#include <QSharedPointer>
#include <QDebug>
#define FROM_PTR
//#define CONST_CREATE
#define FROM_RAW_PTR
#define PERFECT_FORWARD_CREATE
//#define PERFECT_FORWARD_CREATE_CONST
#define BUILTIN_CAST
class A
{
public:
A() = default;
A(int i) : _i{i} {}
void foo() const { qDebug() << "const foo" << _i; }
void foo() { qDebug() << "foo" << ++_i; }
private:
int _i{0};
};
using ASPtr = QSharedPointer<A>;
using ASCPtr = QSharedPointer<const A>;
int main(int argc, char *argv[])
{
Q_UNUSED(argc)
Q_UNUSED(argv)
#ifdef FROM_PTR
qDebug() << "FROM_PTR";
auto a1 = ASPtr::create();
a1->foo();
auto ca1 = static_cast<ASCPtr>(a1);
ca1->foo();
qDebug() << "\n";
#endif // FROM_PTR
#ifdef CONST_CREATE
qDebug() << "CONST_CREATE";
auto ca2 = ASCPtr::create();
ca2->foo();
qDebug() << "\n";
#endif // CONST_CREATE
#ifdef FROM_RAW_PTR
qDebug() << "FROM_RAW_PTR";
auto ca3 = ASCPtr(new const A);
ca3->foo();
qDebug() << "\n";
#endif // FROM_RAW_PTR
#ifdef PERFECT_FORWARD_CREATE
qDebug() << "PERFECT_FORWARD_CREATE";
auto a2 = ASPtr::create(10);
a2->foo();
qDebug() << "\n";
#endif // PERFECT_FORWARD_CREATE
#ifdef PERFECT_FORWARD_CREATE_CONST
qDebug() << "PERFECT_FORWARD_CREATE_CONST";
auto ca4 = ASCPtr::create(20);
ca4->foo();
qDebug() << "\n";
#endif // PERFECT_FORWARD_CREATE
#ifdef BUILTIN_CAST
qDebug() << "BUILTIN_CAST";
QSharedPointer<A> a3 = ASPtr::create();
a3->foo();
auto ca4 = a3.constCast<const A>();
ca4->foo();
qDebug() << "\n";
#endif // BUILTIN_CAST
return 0;
}
That is a known Qt bug (QTBUG-49748). Although it is marked as resolved in Qt 5.6.0, the bug is still present as pointed out in the comments.
Why is this happening?
Look at the implmentation of the class QSharedPointer qsharedpointer_impl.h.
In particular the line:
new (result.data()) T(std::forward<Args>(arguments)...);
uses the result.data() as the new expression placement params. Unfortunately, one can not use a const pointer as a placement param (have a look at this question here on SO for more details).
Hence, there's not much you can do except reporting this to Qt developers via the official bug tracker.
You may have a look at the smart pointers provided by the standard library (e.g. std::shared_ptr) if you are not forced to use Qt ones.
UPDATE
As reported in Qt bug tracker, this bug was fixed in version 5.11 (here is the related commit). Basically, they used std::remove_cv to remove the topmost const from the type specified.
I want to create MPI communicators linking the process with rank 0 to every other process. Suppose n is the total number of processes. Then the process with rank 0 is supposed to have n-1 communicators while each of the other processes has one communicator. Is this possible, and, if it is, why can I not use the program below to achieve this?
Compiling the code below using the mpic++ compiler terminates without warnings and errors on my computer. But when I run the resulting program using 3 or more processes (mpiexec -n 3), it never terminates.
Likely I'm misunderstanding the concept of communicators in MPI. Maybe someone can help me understand why the program below gets stuck, and what is a better way to create those communicators? Thanks.
#include <iostream>
#include <vector>
#include <thread>
#include <chrono>
#include "mpi.h"
void FinalizeMPI();
void InitMPI(int argc, char** argv);
int main(int argc, char** argv) {
InitMPI(argc, argv);
int rank,comm_size;
MPI_Comm_rank(MPI_COMM_WORLD,&rank);
MPI_Comm_size(MPI_COMM_WORLD,&comm_size);
if(comm_size<2) {
FinalizeMPI();
return 0;
}
MPI_Group GroupAll;
MPI_Comm_group(MPI_COMM_WORLD, &GroupAll);
if(rank==0) {
std::vector<MPI_Group> myGroups(comm_size-1);
std::vector<MPI_Comm> myComms(comm_size-1);
for(int k=1;k<comm_size;++k) {
int ranks[2]{0, k};
MPI_Group_incl(GroupAll, 2, ranks, &myGroups[k-1]);
int err = MPI_Comm_create(MPI_COMM_WORLD, myGroups[k-1], &myComms[k-1]);
std::cout << "Error: " << err << std::endl;
}
} else {
MPI_Group myGroup;
MPI_Comm myComm;
int ranks[2]{0,rank};
MPI_Group_incl(GroupAll, 2, ranks, &myGroup);
int err = MPI_Comm_create(MPI_COMM_WORLD, myGroup, &myComm);
std::cout << "Error: " << err << std::endl;
}
std::cout << "Communicators created: " << rank << std::endl;
std::this_thread::sleep_for(std::chrono::milliseconds(1000));
FinalizeMPI();
return 0;
}
void FinalizeMPI() {
int flag;
MPI_Finalized(&flag);
if(!flag)
MPI_Finalize();
}
void InitMPI(int argc, char** argv) {
int flag;
MPI_Initialized(&flag);
if(!flag) {
int provided_Support;
MPI_Init_thread(&argc, &argv, MPI_THREAD_MULTIPLE, &provided_Support);
if(provided_Support!=MPI_THREAD_MULTIPLE) {
exit(0);
}
}
}
MPI_Comm_create is a collective operation on the initial communicator (MPI_COMM_WORLD) - you must call it on all processes.
The simplest way to fix the issue is to use MPI_Comm_create_group just the way you do it. It is similar to MPI_Comm_create except that it is collective over the group.
I'm working with a buffer and I'm trying to get a string from it, but isnt working...
Example:
*void myFunc(QDataStream& in)
{
quint8 v;
in >> v;
// Ok, I caught v value successfuly
QString s;
in >> s;
// Didnt work :<
}*
The string lenght is stored on 2 first bytes...
Thanks
If the string was not written as a QString, you need to read its length and content separately.
quint8 v;
in >> v;
quint16 length = 0;
in >> length;
// the string is probably utf8 or latin
QByteArray buffer(length, Qt::Uninitialized);
in.readRawData(buffer.data(), length);
QString string(buffer);
You might have to change the endianness of the QDataStream with QDataStream::setByteOrder before reading the 16-bit length.
We should really see the writing code and how you create the QDataStream. I tried with the following sample, and in this case your function works very well:
#include <QCoreApplication>
#include <QDebug>
#include <QDataStream>
#include <QBuffer>
void myFunc(QDataStream& in)
{
quint8 v;
in >> v;
qDebug() << v;
// Ok, I caught v value successfuly
QString s;
in >> s;
qDebug() << s;
// Didnt work :<
}
int main(int argc, char ** argv) {
QCoreApplication a(argc, argv);
QBuffer buffer;
buffer.open(QBuffer::ReadWrite);
// write test data into the buffer
QDataStream out(&buffer);
quint8 ival = 42;
QString sval = "Qt";
out << ival;
out << sval;
// read back data
buffer.seek(0);
myFunc(out);
return a.exec();
}
Output when executed:
$ ./App
42
"Qt"