out-of-core TrajectoryIterator
pytraj.trajectory_iterator.TrajectoryIterator(filename=None, top=None, *args, **kwd)¶Bases: pytraj.c_traj.c_trajectory.TrajectoryCpptraj, pytraj.shared_trajectory.SharedTrajectory
out-of-core trajectory holder.
Notes
It’s a bit tricky to pickle this class. As default, new TrajectoryIterator will use original trj filename and top filename. If set _pickle_topology to True, its Topology will be pickled (slow but more accurate if you change the topology in the fly)
Examples
>>> import pytraj as pt
>>> from pytraj.testing import get_fn
>>> traj_name, top_name = get_fn('tz2')
>>> traj = pt.TrajectoryIterator(traj_name, top_name)
>>> # user should always use :method:`pytraj.iterload` to load TrajectoryIterator
>>> traj = pt.iterload(['remd.x.000', 'remd.x.001'], 'input.parm7')
Attributes
filelist |
return a list of input filenames. Order does matter |
filename |
return 1st filename in filelist. For testing only |
metadata |
return a dict of general information |
n_atoms |
used for frame_iter |
n_frames |
total snapshots |
shape |
(n_frames, n_atoms, 3) |
temperatures |
return 1D array of temperatures |
top |
Topology |
topology |
alias of traj.top |
unitcells |
return unitcells (Box) with shape=(n_frames, 6) |
xyz |
return 3D array of coordinates |
Methods
__call__ |
Call self as a function. |
at(index) |
same as traj[index] |
autoimage(self[, command]) |
|
center(self[, command]) |
|
copy() |
return a deep copy. Use this method with care since the copied traj just reuse |
iterchunk([chunksize, start, stop, ...]) |
iterate trajectory by chunk |
iterframe([start, stop, step, mask, ...]) |
iterate trajectory with given frame_indices or given (start, stop, step) |
principal(self, command) |
|
rotate(self, command) |
|
save([filename, overwrite]) |
|
scale(self, command) |
|
strip(mask) |
|
superpose([mask, ref, ref_mask, mass]) |
register to superpose to reference frame when iterating. |
translate(self, command) |
|
view(*args, **kwargs) |
require NGLView |
at(index)¶same as traj[index]
autoimage(self, command='')¶center(self, command='')¶copy()¶return a deep copy. Use this method with care since the copied traj just reuse the filenames
filelist¶return a list of input filenames. Order does matter
filename¶return 1st filename in filelist. For testing only
iterchunk(chunksize=2, start=0, stop=-1, autoimage=False, rmsfit=None)¶iterate trajectory by chunk
| Parameters: | chunk : int, default=2
start : int, default=0 (first frame) stop : int, default=-1 (last frame) autoimage : bool, default=False rmsfit : None | tuple/list of (reference frame, mask) |
|---|
Notes
if using ‘autoimage` with reference frame for rms-fit, make sure to autoimage ref first
Examples
>>> import pytraj as pt
>>> traj = pt.load_sample_data('tz2')
>>> ref = traj[3]
>>> for chunk in traj.iterchunk(3, autoimage=True, rmsfit=(ref, '@CA')): pass
iterframe(start=0, stop=None, step=1, mask=None, autoimage=False, rmsfit=None, copy=False, frame_indices=None)¶iterate trajectory with given frame_indices or given (start, stop, step)
| Parameters: | start : int, default 0 stop : {None, int}, default None
step : int, default 1 mask : {None, str}, default None
autoimage : bool, default False
rmsfit : {None, int, tuple}, default None
frame_indices : {None, array-like}
|
|---|
Examples
>>> import pytraj as pt
>>> traj = pt.load_sample_data('tz2')
>>> for frame in traj.iterframe(0, 8, 2): pass
>>> for frame in traj.iterframe(0, 8, 2, autoimage=True): pass
>>> # use negative index
>>> traj.n_frames
10
>>> fi = traj.iterframe(0, -1, 2, autoimage=True)
>>> fi.n_frames
5
>>> # mask is atom indices
>>> fi = traj.iterframe(0, -1, 2, mask=range(100), autoimage=True)
>>> fi.n_atoms
100
metadata¶return a dict of general information
Examples
>>> traj.metadata
{'box_type': 'ortho',
'has_box': True,
'has_force': False,
'has_replcica_dims': False,
'has_temperature': False,
'has_time': True,
'has_velocity': False,
'n_atoms': 5293,
'n_frames': 10}
n_atoms¶used for frame_iter
n_frames¶total snapshots
principal(self, command)¶rotate(self, command)¶save(filename='', overwrite=False, **kwd)¶scale(self, command)¶shape¶(n_frames, n_atoms, 3)
>>> import pytraj as pt
>>> traj = pt.datafiles.load_tz2_ortho()
>>> traj.shape
(10, 5293, 3)
strip(mask)¶superpose(mask='*', ref=None, ref_mask='', mass=False)¶register to superpose to reference frame when iterating. To turn off superposing, set traj._being_transformed = False
Notes
This method is different from superpose in pytraj.Trajectory.
It does not change the coordinates of TrajectoryCpptraj/TrajectoryIterator itself but
changing the coordinates of copied Frame.
This method is mainly for NGLView in Jupyter notebook, to view out-of-core data. It’s good to do translation and rotation on the fly.
Examples
>>> import pytraj as pt
>>> traj = pt.datafiles.load_tz2()
>>> isinstance(traj, pt.TrajectoryIterator)
True
>>> traj[0].xyz[0]
array([-1.88900006, 9.1590004 , 7.56899977])
>>> # turn on superpose
>>> _ = traj.superpose(ref=-1, mask='@CA')
>>> traj[0].xyz[0]
array([ 6.97324167, 8.82901548, 1.31844696])
>>> # turn off superpose
>>> traj._being_transformed = False
>>> traj[0].xyz[0]
array([-1.88900006, 9.1590004 , 7.56899977])
Example for NGLView:
import pytraj as pt, nglview as nv
traj = pt.datafiles.load_tz2()
traj.superpose(ref=0, mask='@CA')
view = nv.show_pytraj(traj)
view
temperatures¶return 1D array of temperatures
top¶Topology
topology¶alias of traj.top
Examples
>>> import pytraj as pt
>>> from pytraj.testing import get_fn
>>> fname, tname = get_fn('ala3')
>>> traj = pt.iterload(fname, tname)
>>> traj.topology
<Topology: 34 atoms, 3 residues, 1 mols, non-PBC>
>>> new_traj = pt.TrajectoryIterator()
>>> new_traj.topology = traj.topology
>>> new_traj._load(traj.filename)
translate(self, command)¶unitcells¶return unitcells (Box) with shape=(n_frames, 6)
view(*args, **kwargs)¶require NGLView
| Parameters: | args and kwargs : NGLView’s arguments |
|---|
xyz¶return 3D array of coordinates