LCPQ/dft_tools
3
0
Fork 0
mirror of https://github.com/triqs/dft_tools synced 2025-01-12 05:58:18 +01:00
Code Issues Releases Wiki Activity

New notation for Monte Carlo variables

Browse Source
This commit is contained in:
Michel Ferrero 2013-09-09 10:55:54 +02:00
parent fe58c9759f
commit e7d29d8243
4 changed files with 31 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
doc/reference/c++/mctools/ising.rst
View File

@ -168,16 +168,16 @@ The Monte-Carlo itself can now be written::
if (world.rank() == 0) std::cout << "Ising chain" << std::endl; if (world.rank() == 0) std::cout << "Ising chain" << std::endl;
// Prepare the MC parameters // Prepare the MC parameters
int N_Cycles = 500000; int n_cycles = 500000;
int Length_Cycle = 50; int length_cycle = 50;
int N_Warmup_Cycles = 100000; int n_warmup_cycles = 100000;
std::string Random_Name = ""; std::string random_name = "";
int Random_Seed = 374982 + world.rank() * 273894; int random_seed = 374982 + world.rank() * 273894;
int Verbosity = (world.rank() == 0 ? 2 : 0); int verbosity = (world.rank() == 0 ? 2 : 0);
// Construct a Monte Carlo loop // Construct a Monte Carlo loop
triqs::mc_tools::mc_generic<double> IsingMC(N_Cycles, Length_Cycle, N_Warmup_Cycles, triqs::mc_tools::mc_generic<double> IsingMC(n_cycles, length_cycle, n_warmup_cycles,
Random_Name, Random_Seed, Verbosity); random_name, random_seed, verbosity);
// parameters of the model // parameters of the model
int length = 100; int length = 100;

4
doc/reference/c++/mctools/loop.rst
View File

@ -48,8 +48,8 @@ C++ variable names
In the C++ examples, these variables will be called: In the C++ examples, these variables will be called:
* ``N_Cycles`` :math:`= N` * ``n_cycles`` :math:`= N`
* ``Length_Cycle`` :math:`= L` * ``length_cycle`` :math:`= L`
* ``N_Warmup_Cycle`` :math:`= W` * ``N_Warmup_Cycle`` :math:`= W`
You will also have to use these names if you will construct an ``mc_generic`` You will also have to use these names if you will construct an ``mc_generic``

38
doc/reference/c++/mctools/overview.rst
View File

@ -87,16 +87,16 @@ but obviously you would usually want to cut this into pieces for clarity::
if (world.rank() == 0) std::cout << "Isolated spin" << std::endl; if (world.rank() == 0) std::cout << "Isolated spin" << std::endl;
// prepare the MC parameters // prepare the MC parameters
int N_Cycles = 5000000; int n_cycles = 5000000;
int Length_Cycle = 10; int length_cycle = 10;
int N_Warmup_Cycles = 10000; int n_warmup_cycles = 10000;
std::string Random_Name = ""; std::string random_name = "";
int Random_Seed = 374982 + world.rank() * 273894; int random_seed = 374982 + world.rank() * 273894;
int Verbosity = (world.rank() == 0 ? 2 : 0); int verbosity = (world.rank() == 0 ? 2 : 0);
// construct a Monte Carlo loop // construct a Monte Carlo loop
triqs::mc_tools::mc_generic<double> SpinMC(N_Cycles, Length_Cycle, N_Warmup_Cycles, triqs::mc_tools::mc_generic<double> SpinMC(n_cycles, length_cycle, n_warmup_cycles,
Random_Name, Random_Seed, Verbosity); random_name, random_seed, verbosity);
// parameters of the model // parameters of the model
double beta = 0.3; double beta = 0.3;
@ -140,15 +140,15 @@ The lines that follow, define the parameters of the Monte
Carlo simulation and construct a Monte Carlo object Carlo simulation and construct a Monte Carlo object
called ``SpinMC``:: called ``SpinMC``::
int N_Cycles = 5000000; int n_cycles = 5000000;
int Length_Cycle = 10; int length_cycle = 10;
int N_Warmup_Cycles = 10000; int n_warmup_cycles = 10000;
std::string Random_Name = ""; std::string random_name = "";
int Random_Seed = 374982 + world.rank() * 273894; int random_seed = 374982 + world.rank() * 273894;
int Verbosity = (world.rank() == 0 ? 2 : 0); int verbosity = (world.rank() == 0 ? 2 : 0);
triqs::mc_tools::mc_generic<double> SpinMC(N_Cycles, Length_Cycle, N_Warmup_Cycles, triqs::mc_tools::mc_generic<double> SpinMC(n_cycles, length_cycle, n_warmup_cycles,
Random_Name, Random_Seed, Verbosity); random_name, random_seed, verbosity);
The ``SpinMC`` is an instance of the ``mc_generic`` class. First of all, note The ``SpinMC`` is an instance of the ``mc_generic`` class. First of all, note
that you need to include the header ``<triqs/mc_tools/mc_generic.hpp>`` in that you need to include the header ``<triqs/mc_tools/mc_generic.hpp>`` in
@ -161,14 +161,14 @@ number of measurements and the warmup length. The definition of these variables
has been detailed earlier in :ref:`montecarloloop`. has been detailed earlier in :ref:`montecarloloop`.
The next two define the random number generator by giving its name in The next two define the random number generator by giving its name in
``Random_Name`` (an empty string means the default generator, i.e. the Mersenne ``random_name`` (an empty string means the default generator, i.e. the Mersenne
Twister) and the random seed in ``Random_See``. As you see the seed is Twister) and the random seed in ``Random_See``. As you see the seed is
different for all node with the use of ``world.rank()``. different for all node with the use of ``world.rank()``.
Finally, the last parameter sets the verbosity level. 0 means no output, 1 will Finally, the last parameter sets the verbosity level. 0 means no output, 1 will
output the progress level for the current node and 2 additionally shows some output the progress level for the current node and 2 additionally shows some
statistics about the simulation when you call ``collect_results``. As you see, statistics about the simulation when you call ``collect_results``. As you see,
we have put ``Verbosity`` to 2 only for the master node and 0 for all the other we have put ``verbosity`` to 2 only for the master node and 0 for all the other
ones. This way the information will be printed only by the master. ones. This way the information will be printed only by the master.
Moves and measures Moves and measures
@ -269,7 +269,7 @@ Let's look at ``compute_m``::
Here only two methods are expected, ``accumulate`` and ``collect_results``. Here only two methods are expected, ``accumulate`` and ``collect_results``.
The method ``accumulate`` is called every ``Length_Cycle`` Monte Carlo loops. The method ``accumulate`` is called every ``length_cycle`` Monte Carlo loops.
It takes one argument which is the current sign in the Monte Carlo simulation. It takes one argument which is the current sign in the Monte Carlo simulation.
Here, we sum the sign in ``Z`` (the partition function) and the magnetization Here, we sum the sign in ``Z`` (the partition function) and the magnetization
in ``M``. The other method ``collect_results`` is usually called just once at in ``M``. The other method ``collect_results`` is usually called just once at

4
doc/reference/c++/mctools/random.rst
View File

@ -81,8 +81,8 @@ When you construct an instance of a Monte Carlo class ``mc_generic``, this
instance automatically has an access to a random number generator. instance automatically has an access to a random number generator.
Imagine you constructed an instance:: Imagine you constructed an instance::
triqs::mc_tools::mc_generic<double> SpinMC(N_Cycles, Length_Cycle, N_Warmup_Cycles, triqs::mc_tools::mc_generic<double> SpinMC(n_cycles, length_cycle, n_warmup_cycles,
Random_Name, Random_Seed, Verbosity); random_name, random_seed, verbosity);
Now, you can use ``SpinMC`` to have a random number generator:: Now, you can use ``SpinMC`` to have a random number generator::