 |
A simple sampler.
This sampler randomizes the conformation and then uses Monte Carlo and conjugate gradient steps to search for good solutions. Each Monte Carlo move is followed by the specified number of conjugate gradient steps before it is decided whether to accept or reject the move. When tjhe sampler encounters a solution that passes all of the restraint score cutoffs, it will accept the solution and move on to generating another one.
At the moment it only support optimization of Cartesian coordinates, but this will be fixed when people ask for it (and they already have :-). We are also open to supporting a wider variety of optimization protocols (eg only do conjugate gradient steps occasionally).
Examples: chain, basic optimization, nup84 cg, sample 0, nup84 rb
Inheritance diagram for IMP::core::MCCGSampler:Public Member Functions | |
| MCCGSampler (Model *m, std::string name="MCCG Sampler %1%") | |
| ConfigurationSet * | get_rejected_configurations () const |
| void | set_bounding_box (const algebra::BoundingBoxD< 3 > &bb) |
| Set the bounding box for randomizing the Cartesian coordinates. | |
| void | set_is_refining (bool tf) |
| void | set_local_optimizer (Optimizer *opt) |
| Set a local optimizer to use instead of ConjugateGradients. | |
| void | set_max_monte_carlo_step_size (double d) |
| Set the maximum size of the MC step for all attributes. | |
| void | set_max_monte_carlo_step_size (FloatKey k, double d) |
| Set the maximum size of the MC step for an attribute. | |
| void | set_save_rejected_configurations (bool tf) |
| Whether or not to save rejected conformations. | |
Number of steps | |
A sampling run proceeds as 3 nested loops
| |
| void | set_number_of_attempts (unsigned int att) |
| Set the maximum number of attempts to find a solution. | |
| void | set_number_of_monte_carlo_steps (unsigned int cg) |
| Set the number of MC steps to take in each optimization run. | |
| void | set_number_of_conjugate_gradient_steps (unsigned int cg) |
| Set the number of CG steps to take after each MC step. | |
Optimizer states | |
The optimizer states will be added to the MonteCarlo optimizer used. | |
| void | remove_optimizer_state (OptimizerState *d) |
| template<class F > | |
| void | remove_optimizer_states_if (const F &f) |
| template<class List > | |
| void | remove_optimizer_states (List d) |
| template<class List > | |
| void | set_optimizer_states (List ps) |
| template<class List > | |
| void | set_optimizer_states_order (List ps) |
| unsigned int | add_optimizer_state (OptimizerState *obj) |
| template<class List > | |
| void | add_optimizer_states (List objs) |
| void | clear_optimizer_states () |
| unsigned int | get_number_of_optimizer_states () const |
| bool | get_has_optimizer_states () const |
| OptimizerState * | get_optimizer_state (unsigned int i) const |
| OptimizerStates | get_optimizer_states () const |
| void | reserve_optimizer_states (unsigned int sz) |
| OptimizerStateIterator | optimizer_states_begin () |
| OptimizerStateIterator | optimizer_states_end () |
| OptimizerStateConstIterator | optimizer_states_begin () const |
| OptimizerStateConstIterator | optimizer_states_end () const |
| static MCCGSampler * | get_from (IMP::base::Object *o) |
| void IMP::core::MCCGSampler::set_bounding_box | ( | const algebra::BoundingBoxD< 3 > & | bb | ) |
Set the bounding box for randomizing the Cartesian coordinates.
| void IMP::core::MCCGSampler::set_is_refining | ( | bool | tf | ) |
if set to true, then do not randomize the configuration before sampling.
| void IMP::core::MCCGSampler::set_local_optimizer | ( | Optimizer * | opt | ) |
Set a local optimizer to use instead of ConjugateGradients.
| void IMP::core::MCCGSampler::set_max_monte_carlo_step_size | ( | double | d | ) |
Set the maximum size of the MC step for all attributes.
| void IMP::core::MCCGSampler::set_max_monte_carlo_step_size | ( | FloatKey | k, |
| double | d | ||
| ) |
Set the maximum size of the MC step for an attribute.
As was mentioned, at the moment k can be one of x,y or z.
| void IMP::core::MCCGSampler::set_number_of_attempts | ( | unsigned int | att | ) |
Set the maximum number of attempts to find a solution.
| void IMP::core::MCCGSampler::set_number_of_conjugate_gradient_steps | ( | unsigned int | cg | ) |
Set the number of CG steps to take after each MC step.
| void IMP::core::MCCGSampler::set_number_of_monte_carlo_steps | ( | unsigned int | cg | ) |
Set the number of MC steps to take in each optimization run.
| void IMP::core::MCCGSampler::set_save_rejected_configurations | ( | bool | tf | ) |
Whether or not to save rejected conformations.
Saving these can be useful if the sampling is not finding any good conformations.
