Difference between revisions of "Tutorials:Parameters"
m |
|||
(6 intermediate revisions by 3 users not shown) | |||
Line 1: | Line 1: | ||
+ | {{Languages|Tutorials:Parameters}} | ||
+ | |||
+ | |||
The following input parameters are common to most of the ALPS applications | The following input parameters are common to most of the ALPS applications | ||
Line 78: | Line 81: | ||
| align="center"| 0 | | align="center"| 0 | ||
|| The random number seed used in the next run. After using a seed in the creation of a Monte Carlo run, this value gets incremented by one. | || The random number seed used in the next run. After using a seed in the creation of a Monte Carlo run, this value gets incremented by one. | ||
+ | |- | ||
+ | || RNG | ||
+ | | align="center"| "mt19937" | ||
+ | || The pseudo-random number generator to be used. Allowed values are [http://www.boost.org/doc/libs/1_35_0/libs/random/random-generators.html#lagged_fibonacci_spec "lagged_fibonacci607"] and [http://www.boost.org/doc/libs/1_35_0/libs/random/random-generators.html#mt19937 "mt19937"]. | ||
|- | |- | ||
|| WORK_FACTOR | || WORK_FACTOR | ||
Line 91: | Line 98: | ||
||Number of Monte Carlo sweeps for thermalization | ||Number of Monte Carlo sweeps for thermalization | ||
|} | |} | ||
+ | |||
==Additional parameters for exact diagonalization == | ==Additional parameters for exact diagonalization == | ||
Line 101: | Line 109: | ||
||CONSERVED_QUANTUMNUMBERS | ||CONSERVED_QUANTUMNUMBERS | ||
|| | || | ||
− | ||specifies conserved global quantum numbers which are used to split the computation into smaller computations for the different sectors. If more than one quantum number is conserved, the quantum numbers are listed in double quotes and separated by commas as in CONSERVED_QUANTUMNUMBERS="N,Sz" | + | ||specifies conserved '''global''' quantum numbers which are used to split the computation into smaller computations for the different sectors. If more than one quantum number is conserved, the quantum numbers are listed in double quotes and separated by commas as in CONSERVED_QUANTUMNUMBERS="N,Sz" |
|- | |- | ||
||N_total, Sz_total, ... | ||N_total, Sz_total, ... | ||
Line 109: | Line 117: | ||
||TRANSLATION_SYMMETRY | ||TRANSLATION_SYMMETRY | ||
||true | ||true | ||
− | ||<tt>sparsediag</tt> | + | ||<tt>fulldiag</tt> and <tt>sparsediag</tt> exploit translational symmetry and classify eigenstates by their momentum quantum numbers when possible. This symmetry reductions can be switched off with TRANSLATION_SYMMETRY=false. |
|- | |- | ||
||TOTAL_MOMENTUM | ||TOTAL_MOMENTUM | ||
|| | || | ||
|| fixes the value of the total momentum. Further explanations can be found below. | || fixes the value of the total momentum. Further explanations can be found below. | ||
+ | |- | ||
+ | ||MEASURE_ENERGY | ||
+ | ||false | ||
+ | ||if no measurements are explicitly specified, <tt>fulldiag</tt> and <tt>sparsediag</tt> do not store any information on eigenstates by default. Of course, the energy can always be computed for any eigenstate. If you wish to have this in the output and no other measurements are specified, you can specify MEASURE_ENERGY=true. | ||
|} | |} | ||
+ | '''Note:''' Instead of true and false, you can also specify 1 and 0, respectively. | ||
Line 122: | Line 135: | ||
TOTAL_MOMENTUM takes the momentum quantum numbers as a vector, | TOTAL_MOMENTUM takes the momentum quantum numbers as a vector, | ||
− | + | ''i.e.'' a space-separated list of numbers. Typically, each momentum | |
quantum number <i>k<sub>i</sub></i> will be of the form | quantum number <i>k<sub>i</sub></i> will be of the form | ||
− | + | ||
<center><font size=+1> | <center><font size=+1> | ||
<i>k<sub>i</sub></i> = | <i>k<sub>i</sub></i> = | ||
2 π <i>n<sub>i</sub></i>/<i>L<sub>i</sub></i> , | 2 π <i>n<sub>i</sub></i>/<i>L<sub>i</sub></i> , | ||
</font></center> | </font></center> | ||
− | + | ||
where <i>n<sub>i</sub></i> is an integer and <i>L<sub>i</sub></i> the | where <i>n<sub>i</sub></i> is an integer and <i>L<sub>i</sub></i> the | ||
linear extent in the corresponding direction. It is possible to specify | linear extent in the corresponding direction. It is possible to specify | ||
a symbolic number such as <tt>2*Pi/5</tt> if you put the values in | a symbolic number such as <tt>2*Pi/5</tt> if you put the values in | ||
quotation marks, e.g. <tt>TOTAL_MOMENTUM="2*Pi/5 0"</tt>. | quotation marks, e.g. <tt>TOTAL_MOMENTUM="2*Pi/5 0"</tt>. | ||
− | + | ||
− | + | ||
− | <font color="red"> | + | <font color="red">'''Warning:'''</font> An illegal value of TOTAL_MOMENTUM may lead to incorrect results without any further error message. |
− | TOTAL_MOMENTUM may lead to incorrect results without any further error | ||
− | message. |
Latest revision as of 06:59, 2 March 2012
Languages: |
English • 日本語 (ja) • 繁體中文 (zh-tw) • 简体中文 (zh) |
The following input parameters are common to most of the ALPS applications
Contents
Lattice definition
ALPS applications on lattices specify the lattice with the following three parameters
Parameter | Default | Meaning |
LATTICE_LIBRARY | lattices.xml | path to a file containing lattice descriptions |
LATTICE | name of the lattice, specified by dimensionality, extent and unit cell. | |
GRAPH | as an alternative to a lattice, also a specific arbitrary graph defined in the lattice library can be specified. |
In addition, the lattice description can require further parameters (e.g. L or W) as specified in the lattice description file.
Model definition
ALPS quantum lattice models can be specified using the following parameters
Parameter | Default | Meaning |
MODEL_LIBRARY | models.xml | path to a file containing model descriptions</td> |
MODEL | name of the model (for example "spin" or "boson") |
The model description can also require further parameters (e.g. S=1/2 or S=1, h=0.5 for spin models, t=1.5 or mu=0.5 for boson models) as specified in the model description file.
Parameters for finite temperature simulations
Parameter | Meaning |
T | the temperature |
BETA | inverse of temperature (if temperature is not given |
Additional parameters for Monte Carlo simulations
Parameter | Default | Meaning |
SEED | 0 | The random number seed used in the next run. After using a seed in the creation of a Monte Carlo run, this value gets incremented by one. |
RNG | "mt19937" | The pseudo-random number generator to be used. Allowed values are "lagged_fibonacci607" and "mt19937". |
WORK_FACTOR | 1 | A factor by which the work that needs to be done for a simulation is multiplied in load balancing. |
SWEEPS | number of Monte Carlo steps (after thermalization) | |
THERMALIZATION | Number of Monte Carlo sweeps for thermalization |
Additional parameters for exact diagonalization
Parameter | Default | Meaning |
CONSERVED_QUANTUMNUMBERS | specifies conserved global quantum numbers which are used to split the computation into smaller computations for the different sectors. If more than one quantum number is conserved, the quantum numbers are listed in double quotes and separated by commas as in CONSERVED_QUANTUMNUMBERS="N,Sz" | |
N_total, Sz_total, ... | and similar parameters might be defined for your model through a constraint in your model definition. These constraints will be used if these parameters are specified and the quantumnumber is listed in CONSERVED_QUANTUMNUMBERS. | |
TRANSLATION_SYMMETRY | true | fulldiag and sparsediag exploit translational symmetry and classify eigenstates by their momentum quantum numbers when possible. This symmetry reductions can be switched off with TRANSLATION_SYMMETRY=false. |
TOTAL_MOMENTUM | fixes the value of the total momentum. Further explanations can be found below. | |
MEASURE_ENERGY | false | if no measurements are explicitly specified, fulldiag and sparsediag do not store any information on eigenstates by default. Of course, the energy can always be computed for any eigenstate. If you wish to have this in the output and no other measurements are specified, you can specify MEASURE_ENERGY=true. |
Note: Instead of true and false, you can also specify 1 and 0, respectively.
If the lattice supports translation symmetries, you can specify
the total momentum quantum numbers, but you should be quite careful in
doing so.
TOTAL_MOMENTUM takes the momentum quantum numbers as a vector, i.e. a space-separated list of numbers. Typically, each momentum quantum number k_{i} will be of the form
k_{i} = 2 π n_{i}/L_{i} ,
where n_{i} is an integer and L_{i} the linear extent in the corresponding direction. It is possible to specify a symbolic number such as 2*Pi/5 if you put the values in quotation marks, e.g. TOTAL_MOMENTUM="2*Pi/5 0".