pyemma.msm.BayesianMSM

class pyemma.msm.BayesianMSM(*args, **kwargs)

Bayesian Markov state model estimator

__init__(lag=1, nsamples=100, nsteps=None, reversible=True, statdist_constraint=None, count_mode='effective', sparse=False, connectivity='largest', dt_traj='1 step', conf=0.95, show_progress=True, mincount_connectivity='1/n', core_set=None, milestoning_method='last_core')

Bayesian estimator for MSMs given discrete trajectory statistics

Parameters
  • lag (int, optional, default=1) – lagtime to estimate the HMSM at

  • nsamples (int, optional, default=100) – number of sampled transition matrices used

  • nsteps (int, optional, default=None) – number of Gibbs sampling steps for each transition matrix used. If None, nstep will be determined automatically

  • reversible (bool, optional, default = True) – If true compute reversible MSM, else non-reversible MSM

  • statdist_constraint ((M,) ndarray optional) – Stationary vector on the full set of states. Assign zero stationary probabilities to states for which the stationary vector is unknown. Estimation will be made such that the resulting ensemble of transition matrices is defined on the intersection of the states with positive stationary vector and the largest connected set (undirected).

  • count_mode (str, optional, default='effective') –

    mode to obtain count matrices from discrete trajectories. Should be one of:

    • ’sliding’ : A trajectory of length T will have \(T-\tau\) counts at time indexes .. math:: (0 rightarray tau), (1 rightarray tau+1), …, (T-tau-1 rightarray T-1)

    • ’effective’ : Uses an estimate of the transition counts that are statistically uncorrelated. Recommended when used with a Bayesian MSM.

    • ’sample’ : A trajectory of length T will have \(T / \tau\) counts at time indexes .. math:: (0 rightarray tau), (tau rightarray 2 tau), …, (((T/tau)-1) tau rightarray T)

  • sparse (bool, optional, default = False) – If true compute count matrix, transition matrix and all derived quantities using sparse matrix algebra. In this case python sparse matrices will be returned by the corresponding functions instead of numpy arrays. This behavior is suggested for very large numbers of states (e.g. > 4000) because it is likely to be much more efficient.

  • connectivity (str, optional, default = 'largest') –

    Connectivity mode. Three methods are intended (currently only ‘largest’ is implemented)

    • ’largest’ : The active set is the largest reversibly connected set. All estimation will be done on this subset and all quantities

      (transition matrix, stationary distribution, etc) are only defined on this subset and are correspondingly smaller than the full set of states

    • ’all’ : The active set is the full set of states. Estimation will be conducted on each reversibly connected set separately. That means the transition matrix will decompose into disconnected submatrices, the stationary vector is only defined within subsets, etc. Currently not implemented.

    • ’none’ : The active set is the full set of states. Estimation will be conducted on the full set of states without ensuring connectivity. This only permits nonreversible estimation. Currently not implemented.

  • dt_traj (str, optional, default='1 step') –

    Description of the physical time corresponding to the trajectory time step. May be used by analysis algorithms such as plotting tools to pretty-print the axes. By default ‘1 step’, i.e. there is no physical time unit. Specify by a number, whitespace and unit. Permitted units are (* is an arbitrary string):

    ’fs’, ‘femtosecond*’
    ’ps’, ‘picosecond*’
    ’ns’, ‘nanosecond*’
    ’us’, ‘microsecond*’
    ’ms’, ‘millisecond*’
    ’s’, ‘second*’

  • conf (float, optional, default=0.95) – Confidence interval. By default one-sigma (68.3%) is used. Use 95.4% for two sigma or 99.7% for three sigma.

  • show_progress (bool, default=True) – Show progressbars for calculation?

  • mincount_connectivity (float or '1/n') – minimum number of counts to consider a connection between two states. Counts lower than that will count zero in the connectivity check and may thus separate the resulting transition matrix. The default evaluates to 1/nstates.

  • core_set (None (default) or array like, dtype=int) – Definition of core set for milestoning MSMs. If set to None, replaces state -1 (if found in discrete trajectories) and performs milestone counting. No effect for Voronoi-discretized trajectories (default). If a list or np.ndarray is supplied, discrete trajectories will be assigned accordingly.

  • milestoning_method (str) – Method to use for counting transitions in trajectories with unassigned frames. Currently available: | ‘last_core’, assigns unassigned frames to last visited core

References

1(1,2,3,4)

Trendelkamp-Schroer, B., H. Wu, F. Paul and F. Noe: Estimation and uncertainty of reversible Markov models. J. Chem. Phys. (in review) Preprint: http://arxiv.org/abs/1507.05990

Methods

_Loggable__create_logger()

_ProgressReporterMixin__check_stage_registered(stage)

_SerializableMixIn__interpolate(state, klass)

__delattr__(name, /)

Implement delattr(self, name).

__dir__()

Default dir() implementation.

__eq__(other)

Return self==value.

__format__(format_spec, /)

Default object formatter.

__ge__(value, /)

Return self>=value.

__getattribute__(name, /)

Return getattr(self, name).

__getstate__()

__gt__(value, /)

Return self>value.

__init__([lag, nsamples, nsteps, …])

Bayesian estimator for MSMs given discrete trajectory statistics

__init_subclass__(*args, **kwargs)

This method is called when a class is subclassed.

__le__(value, /)

Return self<=value.

__lt__(value, /)

Return self<value.

__my_getstate__()

__my_setstate__(state)

__ne__(value, /)

Return self!=value.

__new__(cls, *args, **kwargs)

Create and return a new object.

__reduce__()

Helper for pickle.

__reduce_ex__(protocol, /)

Helper for pickle.

__repr__()

Return repr(self).

__setattr__(name, value, /)

Implement setattr(self, name, value).

__setstate__(state)

__sizeof__()

Size of object in memory, in bytes.

__str__()

Return str(self).

__subclasshook__

Abstract classes can override this to customize issubclass().

_assert_in_active(A)

Checks if set A is within the active set

_assert_metastable()

Tests if pcca object is available, or else raises a ValueError.

_blocksplit_dtrajs(dtrajs, sliding)

_check_estimated()

_check_is_estimated()

_check_samples_available()

_cleanup_logger(logger_id, logger_name)

_committor_backward(P, A, B[, mu])

_committor_forward(P, A, B)

_compute_eigendecomposition(neig)

Conducts the eigenvalue decomposition and stores k eigenvalues, left and right eigenvectors

_compute_eigenvalues(neig)

Conducts the eigenvalue decomposition and stores k eigenvalues, left and right eigenvectors

_ensure_eigendecomposition([neig])

Ensures that eigendecomposition has been performed with at least neig eigenpairs

_ensure_eigenvalues([neig])

Ensures that at least neig eigenvalues have been computed

_estimate(dtrajs)

Estimates the MSM

_get_classes_to_inspect()

gets classes self derives from which 1.

_get_dtraj_stats(dtrajs)

Compute raw trajectory counts

_get_interpolation_map(cls)

_get_model_param_names()

Get parameter names for the model

_get_param_names()

Get parameter names for the estimator

_get_private_field(cls, name[, default])

_get_serialize_fields(cls)

_get_state_of_serializeable_fields(klass, state)

:return a dictionary {k:v} for k in self.serialize_fields and v=getattr(self, k)

_get_version(cls[, require])

_get_version_for_class_from_state(state, klass)

retrieves the version of the current klass from the state mapping from old locations to new ones.

_logger_is_active(level)

@param level: int log level (debug=10, info=20, warn=30, error=40, critical=50)

_mfpt(P, A, B[, mu])

_prepare_input_revpi(C, pi)

Max.

_progress_context([stage])

param stage

_progress_force_finish([stage, description])

forcefully finish the progress for given stage

_progress_register(amount_of_work[, …])

Registers a progress which can be reported/displayed via a progress bar.

_progress_set_description(stage, description)

set description of an already existing progress

_progress_update(numerator_increment[, …])

Updates the progress.

_set_state_from_serializeable_fields_and_state(…)

set only fields from state, which are present in klass.__serialize_fields

cktest(nsets[, memberships, mlags, conf, …])

Conducts a Chapman-Kolmogorow test.

coarse_grain(ncoarse[, method])

Returns a coarse-grained Markov model.

committor_backward(A, B)

Backward committor from set A to set B

committor_forward(A, B)

Forward committor (also known as p_fold or splitting probability) from set A to set B

correlation(a[, b, maxtime, k, ncv])

Time-correlation for equilibrium experiment.

eigenvalues([k])

Compute the transition matrix eigenvalues

eigenvectors_left([k])

Compute the left transition matrix eigenvectors

eigenvectors_right([k])

Compute the right transition matrix eigenvectors

estimate(dtrajs, **kw)

param dtrajs

discrete trajectories, stored as integer ndarrays (arbitrary size)

expectation(a)

Equilibrium expectation value of a given observable.

fingerprint_correlation(a[, b, k, ncv])

Dynamical fingerprint for equilibrium time-correlation experiment.

fingerprint_relaxation(p0, a[, k, ncv])

Dynamical fingerprint for perturbation/relaxation experiment.

fit(X[, y])

Estimates parameters - for compatibility with sklearn.

generate_traj(N[, start, stop, stride])

Generates a synthetic discrete trajectory of length N and simulation time stride * lag time * N

get_model_params([deep])

Get parameters for this model.

get_params([deep])

Get parameters for this estimator.

hmm(nhidden)

Estimates a hidden Markov state model as described in 1

load(file_name[, model_name])

Loads a previously saved PyEMMA object from disk.

mfpt(A, B)

Mean first passage times from set A to set B, in units of the input trajectory time step

pcca(m)

Runs PCCA++ 1 to compute a metastable decomposition of MSM states

propagate(p0, k)

Propagates the initial distribution p0 k times

relaxation(p0, a[, maxtime, k, ncv])

Simulates a perturbation-relaxation experiment.

sample_by_distributions(distributions, nsample)

Generates samples according to given probability distributions

sample_by_state(nsample[, subset, replace])

Generates samples of the connected states.

sample_conf(f, *args, **kwargs)

Sample confidence interval of numerical method f over all samples

sample_f(f, *args, **kwargs)

Evaluated method f for all samples

sample_mean(f, *args, **kwargs)

Sample mean of numerical method f over all samples

sample_std(f, *args, **kwargs)

Sample standard deviation of numerical method f over all samples

save(file_name[, model_name, overwrite, …])

saves the current state of this object to given file and name.

score(dtrajs[, score_method, score_k])

Scores the MSM using the dtrajs using the variational approach for Markov processes 1 [2]_

score_cv(dtrajs[, n, score_method, score_k])

Scores the MSM using the variational approach for Markov processes 1 [2]_ and crossvalidation [3]_ .

set_model_params([samples, conf, P, pi, …])

param samples

sampled MSMs

set_params(**params)

Set the parameters of this estimator.

simulate(N[, start, stop, dt])

Generates a realization of the Markov Model

timescales([k])

The relaxation timescales corresponding to the eigenvalues

trajectory_weights()

Uses the MSM to assign a probability weight to each trajectory frame.

update_model_params(**params)

Update given model parameter if they are set to specific values

Attributes

P

The transition matrix on the active set.

_BayesianMSM__serialize_version

_Estimator__serialize_fields

_Loggable__ids

_Loggable__refs

_MSMEstimator__serialize_fields

_MSMEstimator__serialize_version

_MSM__serialize_fields

_MSM__serialize_version

_MaximumLikelihoodMSM__serialize_fields

_MaximumLikelihoodMSM__serialize_version

_SampledMSM__serialize_version

_SerializableMixIn__serialize_fields

_SerializableMixIn__serialize_modifications_map

_SerializableMixIn__serialize_version

__dict__

__doc__

__hash__

__module__

__weakref__

list of weak references to the object (if defined)

_estimated

_loglevel_CRITICAL

_loglevel_DEBUG

_loglevel_ERROR

_loglevel_INFO

_loglevel_WARN

_pg_threshold

_prog_rep_callbacks

_prog_rep_descriptions

_prog_rep_progressbars

_progress_num_registered

_progress_registered_stages

_save_data_producer

active_count_fraction

The fraction of counts in the largest connected set.

active_set

The active set of states on which all computations and estimations will be done

active_state_fraction

The fraction of states in the largest connected set.

active_state_indexes

Ensures that the connected states are indexed and returns the indices

connected_sets

The reversible connected sets of states, sorted by size (descending)

connectivity

Returns the connectivity mode of the MSM

core_set

list of states which are defined to lie within the core set.

count_matrix_active

The count matrix on the active set given the connectivity mode used.

count_matrix_full

The count matrix on full set of discrete states, irrespective as to whether they are connected or not.

discrete_trajectories_active

A list of integer arrays with the discrete trajectories mapped to the connectivity mode used.

discrete_trajectories_full

A list of integer arrays with the original (unmapped) discrete trajectories:

discrete_trajectories_unmodified

A list of integer arrays with the original and not modified discrete trajectories.

dt_model

Description of the physical time corresponding to the lag.

dt_traj

dtrajs_active

A list of integer arrays with the discrete trajectories mapped to the connectivity mode used.

dtrajs_full

A list of integer arrays with the original (unmapped) discrete trajectories:

dtrajs_milestone_counting_offsets

Offsets for milestone counted trajectories for each input discrete trajectory.

dtrajs_unmodified

A list of integer arrays with the original and not modified discrete trajectories.

effective_count_matrix

Statistically uncorrelated transition counts within the active set of states

is_reversible

Returns whether the MSM is reversible

is_sparse

Returns whether the MSM is sparse

lag

The lag time at which the Markov model was estimated

lagtime

The lag time at which the Markov model was estimated

largest_connected_set

The largest reversible connected set of states

logger

The logger for this class instance

metastable_assignments

Assignment of states to metastable sets using PCCA++

metastable_distributions

Probability of metastable states to visit an MSM state by PCCA++

metastable_memberships

Probabilities of MSM states to belong to a metastable state by PCCA++

metastable_sets

Metastable sets using PCCA++

model

The model estimated by this Estimator

n_jobs

Returns number of jobs/threads to use during assignment of data.

n_metastable

Number of states chosen for PCCA++ computation.

name

The name of this instance

neig

number of eigenvalues to compute.

nstates

Number of active states on which all computations and estimations are done

nstates_full

Number of states in discrete trajectories

pi

The stationary distribution on the MSM states

reversible

Returns whether the MSM is reversible

samples

show_progress

whether to show the progress of heavy calculations on this object.

sparse

Returns whether the MSM is sparse

stationary_distribution

The stationary distribution on the MSM states

timestep_model

Physical time corresponding to one transition matrix step, e.g.

transition_matrix

The transition matrix on the active set.