def n_jobs(self, val):
""" set number of jobs/threads to use via assignment of data.
Parameters
----------
val: int or None
a positive int for the number of jobs. Or None to usage all available resources.
Notes
-----
"""
from pyemma.util.reflection import get_default_args
def_args = get_default_args(self.__init__)
# default value from constructor?
if val == def_args['n_jobs']:
omp_threads_from_env = os.getenv('OMP_NUM_THREADS', None)
import psutil
n_cpus = psutil.cpu_count()
if omp_threads_from_env:
try:
self._n_jobs = int(omp_threads_from_env)
self.logger.info(
"number of threads obtained from env variable"
" 'OMP_NUM_THREADS'=%s" % omp_threads_from_env)
except ValueError as ve:
self.logger.warning(
"could not parse env variable 'OMP_NUM_THREADS'."
" Value='{}'. Error={}. Will use {} jobs.".format(
omp_threads_from_env, ve, n_cpus))
self._n_jobs = n_cpus
else:
self._n_jobs = n_cpus
else:
self._n_jobs = int(val)
def __init__(self, dim=-1, var_cutoff=0.95, mean=None):
r""" Principal component analysis.
Given a sequence of multivariate data :math:`X_t`,
computes the mean-free covariance matrix.
.. math:: C = (X - \mu)^T (X - \mu)
and solves the eigenvalue problem
.. math:: C r_i = \sigma_i r_i,
where :math:`r_i` are the principal components and :math:`\sigma_i` are
their respective variances.
When used as a dimension reduction method, the input data is projected onto
the dominant principal components.
Parameters
----------
dim : int, optional, default -1
the number of dimensions (independent components) to project onto. A call to the
:func:`map <pyemma.coordinates.transform.TICA.map>` function reduces the d-dimensional
input to only dim dimensions such that the data preserves the maximum possible autocorrelation
amongst dim-dimensional linear projections.
-1 means all numerically available dimensions will be used unless reduced by var_cutoff.
Setting dim to a positive value is exclusive with var_cutoff.
var_cutoff : float in the range [0,1], optional, default 0.95
Determines the number of output dimensions by including dimensions until their cumulative kinetic variance
exceeds the fraction subspace_variance. var_cutoff=1.0 means all numerically available dimensions
(see epsilon) will be used, unless set by dim. Setting var_cutoff smaller than 1.0 is exclusive with dim
mean : ndarray, optional, default None
Optionally pass pre-calculated means to avoid their re-computation.
The shape has to match the input dimension.
"""
super(PCA, self).__init__()
self._dim = dim
self._var_cutoff = var_cutoff
default_var_cutoff = get_default_args(self.__init__)['var_cutoff']
if dim != -1 and var_cutoff != default_var_cutoff:
raise ValueError('Trying to set both the number of dimension and the subspace variance. Use either or.')
self._dot_prod_tmp = None
self.Y = None
self._N_mean = 0
self._N_cov = 0
self.mu = mean
# set up result variables
self.eigenvalues = None
self.eigenvectors = None
self.cumvar = None
# output options
self._custom_param_progress_handling = True
def __init__(self, dim=-1, var_cutoff=0.95, mean=None, stride=1, skip=0):
r""" Principal component analysis.
.. deprecated:: 2.5.11
Use the scikit-learn
`PCA <https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html>`__
implementation instead. Will be removed in PyEMMA 3.
Given a sequence of multivariate data :math:`X_t`,
computes the mean-free covariance matrix.
.. math:: C = (X - \mu)^T (X - \mu)
and solves the eigenvalue problem
.. math:: C r_i = \sigma_i r_i,
where :math:`r_i` are the principal components and :math:`\sigma_i` are
their respective variances.
When used as a dimension reduction method, the input data is projected onto
the dominant principal components.
Parameters
----------
dim : int, optional, default -1
the number of dimensions (independent components) to project onto. A call to the
:func:`map <pyemma.coordinates.transform.TICA.map>` function reduces the d-dimensional
input to only dim dimensions such that the data preserves the maximum possible autocorrelation
amongst dim-dimensional linear projections.
-1 means all numerically available dimensions will be used unless reduced by var_cutoff.
Setting dim to a positive value is exclusive with var_cutoff.
var_cutoff : float in the range [0,1], optional, default 0.95
Determines the number of output dimensions by including dimensions until their cumulative kinetic variance
exceeds the fraction subspace_variance. var_cutoff=1.0 means all numerically available dimensions
(see epsilon) will be used, unless set by dim. Setting var_cutoff smaller than 1.0 is exclusive with dim
mean : ndarray, optional, default None
Optionally pass pre-calculated means to avoid their re-computation.
The shape has to match the input dimension.
skip: int, default 0
skip the first n frames of each trajectory.
"""
super(PCA, self).__init__()
default_var_cutoff = get_default_args(self.__init__)['var_cutoff']
if dim != -1 and var_cutoff != default_var_cutoff:
raise ValueError('Trying to set both the number of dimension and the subspace variance. Use either or.')
self._model = PCAModel()
self.set_params(dim=dim, var_cutoff=var_cutoff, mean=mean, stride=stride, skip=skip)
def __init__(self,
lag,
dim=-1,
var_cutoff=0.95,
kinetic_map=True,
commute_map=False,
epsilon=1e-6,
stride=1,
skip=0,
reversible=True,
weights=None,
ncov_max=float('inf')):
r""" Time-lagged independent component analysis (TICA) [1]_, [2]_, [3]_.
Parameters
----------
lag : int
lag time
dim : int, optional, default -1
Maximum number of significant independent components to use to reduce dimension of input data. -1 means
all numerically available dimensions (see epsilon) will be used unless reduced by var_cutoff.
Setting dim to a positive value is exclusive with var_cutoff.
var_cutoff : float in the range [0,1], optional, default 0.95
Determines the number of output dimensions by including dimensions until their cumulative kinetic variance
exceeds the fraction subspace_variance. var_cutoff=1.0 means all numerically available dimensions
(see epsilon) will be used, unless set by dim. Setting var_cutoff smaller than 1.0 is exclusive with dim
kinetic_map : bool, optional, default True
Eigenvectors will be scaled by eigenvalues. As a result, Euclidean distances in the transformed data
approximate kinetic distances [4]_. This is a good choice when the data is further processed by clustering.
commute_map : bool, optional, default False
Eigenvector_i will be scaled by sqrt(timescale_i / 2). As a result, Euclidean distances in the transformed
data will approximate commute distances [5]_.
epsilon : float
eigenvalue norm cutoff. Eigenvalues of C0 with norms <= epsilon will be
cut off. The remaining number of eigenvalues define the size
of the output.
stride: int, optional, default = 1
Use only every stride-th time step. By default, every time step is used.
skip : int, default=0
skip the first initial n frames per trajectory.
reversible: bool, default=True
symmetrize correlation matrices C_0, C_{\tau}. At the moment, setting reversible=False is not implemented.
weights: object, optional, default = None
An object that allows to compute re-weighting factors to estimate equilibrium means and correlations from
off-equilibrium data. The only requirement is that weights possesses a method weights(X), that accepts a
trajectory X (np.ndarray(T, n)) and returns a vector of re-weighting factors (np.ndarray(T,)).
Notes
-----
Given a sequence of multivariate data :math:`X_t`, computes the mean-free
covariance and time-lagged covariance matrix:
.. math::
C_0 &= (X_t - \mu)^T (X_t - \mu) \\
C_{\tau} &= (X_t - \mu)^T (X_{t + \tau} - \mu)
and solves the eigenvalue problem
.. math:: C_{\tau} r_i = C_0 \lambda_i(tau) r_i,
where :math:`r_i` are the independent components and :math:`\lambda_i(tau)` are
their respective normalized time-autocorrelations. The eigenvalues are
related to the relaxation timescale by
.. math:: t_i(tau) = -\tau / \ln |\lambda_i|.
When used as a dimension reduction method, the input data is projected
onto the dominant independent components.
References
----------
.. [1] Perez-Hernandez G, F Paul, T Giorgino, G De Fabritiis and F Noe. 2013.
Identification of slow molecular order parameters for Markov model construction
J. Chem. Phys. 139, 015102. doi:10.1063/1.4811489
.. [2] Schwantes C, V S Pande. 2013.
Improvements in Markov State Model Construction Reveal Many Non-Native Interactions in the Folding of NTL9
J. Chem. Theory. Comput. 9, 2000-2009. doi:10.1021/ct300878a
.. [3] L. Molgedey and H. G. Schuster. 1994.
Separation of a mixture of independent signals using time delayed correlations
Phys. Rev. Lett. 72, 3634.
.. [4] Noe, F. and Clementi, C. 2015. Kinetic distance and kinetic maps from molecular dynamics simulation.
J. Chem. Theory. Comput. doi:10.1021/acs.jctc.5b00553
.. [5] Noe, F., Banisch, R., Clementi, C. 2016. Commute maps: separating slowly-mixing molecular configurations
for kinetic modeling. J. Chem. Theory. Comput. doi:10.1021/acs.jctc.6b00762
"""
default_var_cutoff = get_default_args(self.__init__)['var_cutoff']
if dim != -1 and var_cutoff != default_var_cutoff:
raise ValueError(
'Trying to set both the number of dimension and the subspace variance. Use either or.'
)
if kinetic_map and commute_map:
raise ValueError(
'Trying to use both kinetic_map and commute_map. Use either or.'
)
if not reversible:
raise NotImplementedError(
"Reversible=False is currently not implemented.")
# if (kinetic_map or commute_map) and not reversible:
# raise NotImplementedError('kinetic_map and commute_map are not yet implemented for irreversible processes.')
super(TICA, self).__init__()
if dim > -1:
var_cutoff = 1.0
self._covar = LaggedCovariance(c00=True,
c0t=True,
ctt=False,
remove_data_mean=True,
reversible=reversible,
lag=lag,
bessel=False,
stride=stride,
skip=skip,
weights=weights,
ncov_max=ncov_max)
# empty dummy model instance
self._model = TICAModel()
self.set_params(lag=lag,
dim=dim,
var_cutoff=var_cutoff,
kinetic_map=kinetic_map,
commute_map=commute_map,
epsilon=epsilon,
reversible=reversible,
stride=stride,
skip=skip,
weights=weights,
ncov_max=ncov_max)
def __init__(self,
lag,
dim=-1,
var_cutoff=0.95,
kinetic_map=True,
epsilon=1e-6,
force_eigenvalues_le_one=False,
mean=None):
r""" Time-lagged independent component analysis (TICA) [1]_, [2]_, [3]_.
Parameters
----------
tau : int
lag time
dim : int, optional, default -1
Maximum number of significant independent components to use to reduce dimension of input data. -1 means
all numerically available dimensions (see epsilon) will be used unless reduced by var_cutoff.
Setting dim to a positive value is exclusive with var_cutoff.
var_cutoff : float in the range [0,1], optional, default 0.95
Determines the number of output dimensions by including dimensions until their cumulative kinetic variance
exceeds the fraction subspace_variance. var_cutoff=1.0 means all numerically available dimensions
(see epsilon) will be used, unless set by dim. Setting var_cutoff smaller than 1.0 is exclusive with dim
kinetic_map : bool, optional, default True
Eigenvectors will be scaled by eigenvalues. As a result, Euclidean distances in the transformed data
approximate kinetic distances [4]_. This is a good choice when the data is further processed by clustering.
epsilon : float
eigenvalue norm cutoff. Eigenvalues of C0 with norms <= epsilon will be
cut off. The remaining number of eigenvalues define the size
of the output.
force_eigenvalues_le_one : boolean
Compute covariance matrix and time-lagged covariance matrix such
that the generalized eigenvalues are always guaranteed to be <= 1.
mean : ndarray, optional, default None
Optionally pass pre-calculated means to avoid their re-computation.
The shape has to match the input dimension.
Notes
-----
Given a sequence of multivariate data :math:`X_t`, computes the mean-free
covariance and time-lagged covariance matrix:
.. math::
C_0 &= (X_t - \mu)^T (X_t - \mu) \\
C_{\tau} &= (X_t - \mu)^T (X_{t + \tau} - \mu)
and solves the eigenvalue problem
.. math:: C_{\tau} r_i = C_0 \lambda_i(tau) r_i,
where :math:`r_i` are the independent components and :math:`\lambda_i(tau)` are
their respective normalized time-autocorrelations. The eigenvalues are
related to the relaxation timescale by
.. math:: t_i(tau) = -\tau / \ln |\lambda_i|.
When used as a dimension reduction method, the input data is projected
onto the dominant independent components.
References
----------
.. [1] Perez-Hernandez G, F Paul, T Giorgino, G De Fabritiis and F Noe. 2013.
Identification of slow molecular order parameters for Markov model construction
J. Chem. Phys. 139, 015102. doi:10.1063/1.4811489
.. [2] Schwantes C, V S Pande. 2013.
Improvements in Markov State Model Construction Reveal Many Non-Native Interactions in the Folding of NTL9
J. Chem. Theory. Comput. 9, 2000-2009. doi:10.1021/ct300878a
.. [3] L. Molgedey and H. G. Schuster. 1994.
Separation of a mixture of independent signals using time delayed correlations
Phys. Rev. Lett. 72, 3634.
.. [4] Noe, F. and C. Clementi. 2015.
Kinetic distance and kinetic maps from molecular dynamics simulation
http://arxiv.org/abs/1506.06259
"""
super(TICA, self).__init__()
# store lag time to set it appropriately in second pass of parametrize
self._lag = lag
self._dim = dim
self._var_cutoff = var_cutoff
default_var_cutoff = get_default_args(self.__init__)['var_cutoff']
if dim != -1 and var_cutoff != default_var_cutoff:
raise ValueError(
'Trying to set both the number of dimension and the subspace variance. Use either or.'
)
self._kinetic_map = kinetic_map
self._epsilon = epsilon
self._force_eigenvalues_le_one = force_eigenvalues_le_one
# covariances
self.cov = None
self.cov_tau = None
# mean
self.mu = mean
self._N_mean = 0
self._N_cov = 0
self._N_cov_tau = 0
self._eigenvalues = None
self._eigenvectors = None
self._cumvar = None
self._custom_param_progress_handling = True
# skipped trajectories
self._skipped_trajs = []