Elements and Lattices¶
PyTao provides Pydantic models that let you query Tao element data into structured Python objects.
The two main classes are Element (a single lattice element) and Lattice (a collection of elements).
Setup¶
In [1]:
Copied!
from pytao import Tao
tao = Tao(init_file="$ACC_ROOT_DIR/bmad-doc/tao_examples/cesr/tao.init", plot="mpl")
from pytao import Tao
tao = Tao(init_file="$ACC_ROOT_DIR/bmad-doc/tao_examples/cesr/tao.init", plot="mpl")
Getting a single Element¶
Use tao.ele() to get one element. By default, it loads a comprehensive set of data (twiss, orbit, attributes, floor coordinates, etc.).
In [2]:
Copied!
ele = tao.ele("Q00W")
ele.name, ele.key
ele = tao.ele("Q00W")
ele.name, ele.key
Out[2]:
('Q00W', 'Quadrupole')
Head information¶
Every element has a head with metadata: name, key (element type), position, indices, and capability flags.
In [3]:
Copied!
ele.head
ele.head
Out[3]:
ElementHead(
alias='Q00W',
descrip='',
has_ab_multipoles=True,
has_ac_kick=False,
has_control=False,
has_floor=True,
has_kt_multipoles=False,
has_lord_slave=True,
has_mat6=True,
has_methods=True,
has_multipoles_elec=True,
has_photon=False,
has_spin_taylor=False,
has_taylor=False,
has_twiss=True,
has_wake=False,
has_wall3d=0,
is_on=True,
ix_branch=0,
ix_ele=870,
key='Quadrupole',
name='Q00W',
num_cartesian_map=0,
num_cylindrical_map=0,
num_gen_grad_map=0,
num_grid_field=0,
ref_time=7.21417751641456e-09,
s=2.16275600000009,
s_start=0.637956,
type='',
universe=1
)
General attributes¶
General attributes are element-type-specific parameters like L (length), k1 (quadrupole strength), etc.
In [4]:
Copied!
ele.attribs
ele.attribs
Out[4]:
{'L': 1.524800000000091,
'tilt': 0.0754952,
'k1': -0.8417848364530158,
'fringe_type': 'None',
'fringe_at': 'Both_Ends',
'spin_fringe_on': True,
'r0_elec': 0.0,
'r0_mag': 0.0,
'fq1': 0.0,
'fq2': 0.0,
'x_pitch': 0.0,
'y_pitch': 0.0,
'x_offset': 0.0,
'y_offset': 0.0,
'z_offset': 0.0,
'hkick': 0.0,
'vkick': 0.0,
'bl_hkick': 0.0,
'bl_vkick': 0.0,
'b1_gradient': 14.85094057709878,
'ptc_canonical_coords': True,
'delta_ref_time': 5.086185347320325e-09,
'p0c': 5288999975.314811,
'e_tot': 5289000000.000001,
'x_pitch_tot': 0.0,
'y_pitch_tot': 0.0,
'x_offset_tot': 0.0,
'y_offset_tot': 0.0,
'z_offset_tot': 0.0,
'tilt_tot': 0.0754952,
'ref_time_start': 2.127992169094236e-09,
'integrator_order': 2,
'num_steps': 64,
'ds_step': 0.02382500000000142,
'csr_ds_step': 0.0,
'lord_pad1': 0.0,
'lord_pad2': 0.0,
'x1_limit': 0.0,
'x2_limit': 0.0,
'y1_limit': 0.0,
'y2_limit': 0.0,
'lord_status': 'Super_Lord',
'slave_status': 'Minor_Slave',
'aperture_at': 'Exit_End',
'offset_moves_aperture': False,
'aperture_type': 'Rectangular',
'field_master': False}
In [5]:
Copied!
# Access general attributes which are element-specific:
ele.attribs["L"], ele.attribs.lord_status, ele.attribs.aperture_type
# Access general attributes which are element-specific:
ele.attribs["L"], ele.attribs.lord_status, ele.attribs.aperture_type
Out[5]:
(1.524800000000091, 'Super_Lord', 'Rectangular')
You can also view units for these attributes like so:
In [6]:
Copied!
ele.attrs["L"].units
ele.attrs["L"].units
Out[6]:
'm'
The full set of attribute information provided by Tao can be inspected in ele.attrs:
In [7]:
Copied!
ele.attrs
ele.attrs
Out[7]:
GeneralAttributes(
attrs={
'L': Attr(
name='L',
type='REAL',
settable=False,
data=1.524800000000091,
units='m'
),
'tilt': Attr(
name='TILT',
type='REAL',
settable=False,
data=0.0754952,
units='rad'
),
'k1': Attr(
name='K1',
type='REAL',
settable=False,
data=-0.8417848364530158,
units='1/m^2'
),
'fringe_type': Attr(
name='FRINGE_TYPE',
type='ENUM',
settable=False,
data='None',
units=None
),
'fringe_at': Attr(
name='FRINGE_AT',
type='ENUM',
settable=False,
data='Both_Ends',
units=None
),
'spin_fringe_on': Attr(
name='SPIN_FRINGE_ON',
type='LOGIC',
settable=False,
data=True,
units=None
),
'r0_elec': Attr(
name='R0_ELEC',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'r0_mag': Attr(
name='R0_MAG',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'fq1': Attr(
name='FQ1',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'fq2': Attr(
name='FQ2',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'x_pitch': Attr(
name='X_PITCH',
type='REAL',
settable=False,
data=0.0,
units='rad'
),
'y_pitch': Attr(
name='Y_PITCH',
type='REAL',
settable=False,
data=0.0,
units='rad'
),
'x_offset': Attr(
name='X_OFFSET',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'y_offset': Attr(
name='Y_OFFSET',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'z_offset': Attr(
name='Z_OFFSET',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'hkick': Attr(
name='HKICK',
type='REAL',
settable=False,
data=0.0,
units=''
),
'vkick': Attr(
name='VKICK',
type='REAL',
settable=False,
data=0.0,
units=''
),
'bl_hkick': Attr(
name='BL_HKICK',
type='REAL',
settable=False,
data=0.0,
units='T*m'
),
'bl_vkick': Attr(
name='BL_VKICK',
type='REAL',
settable=False,
data=0.0,
units='T*m'
),
'b1_gradient': Attr(
name='B1_GRADIENT',
type='REAL',
settable=False,
data=14.85094057709878,
units='T/m'
),
'ptc_canonical_coords': Attr(
name='PTC_CANONICAL_COORDS',
type='LOGIC',
settable=False,
data=True,
units=None
),
'delta_ref_time': Attr(
name='DELTA_REF_TIME',
type='REAL',
settable=False,
data=5.086185347320325e-09,
units='sec'
),
'p0c': Attr(
name='P0C',
type='REAL',
settable=False,
data=5288999975.314811,
units='eV'
),
'e_tot': Attr(
name='E_TOT',
type='REAL',
settable=False,
data=5289000000.000001,
units='eV'
),
'x_pitch_tot': Attr(
name='X_PITCH_TOT',
type='REAL',
settable=False,
data=0.0,
units='rad'
),
'y_pitch_tot': Attr(
name='Y_PITCH_TOT',
type='REAL',
settable=False,
data=0.0,
units='rad'
),
'x_offset_tot': Attr(
name='X_OFFSET_TOT',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'y_offset_tot': Attr(
name='Y_OFFSET_TOT',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'z_offset_tot': Attr(
name='Z_OFFSET_TOT',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'tilt_tot': Attr(
name='TILT_TOT',
type='REAL',
settable=False,
data=0.0754952,
units='rad'
),
'ref_time_start': Attr(
name='REF_TIME_START',
type='REAL',
settable=False,
data=2.127992169094236e-09,
units='sec'
),
'integrator_order': Attr(
name='INTEGRATOR_ORDER',
type='INT',
settable=False,
data=2,
units=None
),
'num_steps': Attr(
name='NUM_STEPS',
type='INT',
settable=False,
data=64,
units=None
),
'ds_step': Attr(
name='DS_STEP',
type='REAL',
settable=False,
data=0.02382500000000142,
units='m'
),
'csr_ds_step': Attr(
name='CSR_DS_STEP',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'lord_pad1': Attr(
name='LORD_PAD1',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'lord_pad2': Attr(
name='LORD_PAD2',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'x1_limit': Attr(
name='X1_LIMIT',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'x2_limit': Attr(
name='X2_LIMIT',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'y1_limit': Attr(
name='Y1_LIMIT',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'y2_limit': Attr(
name='Y2_LIMIT',
type='REAL',
settable=False,
data=0.0,
units='m'
),
'lord_status': Attr(
name='lord_status',
type='ENUM',
settable=False,
data='Super_Lord',
units=None
),
'slave_status': Attr(
name='slave_status',
type='ENUM',
settable=False,
data='Minor_Slave',
units=None
),
'aperture_at': Attr(
name='aperture_at',
type='ENUM',
settable=True,
data='Exit_End',
units=None
),
'offset_moves_aperture': Attr(
name='offset_moves_aperture',
type='LOGIC',
settable=True,
data=False,
units=None
),
'aperture_type': Attr(
name='aperture_type',
type='ENUM',
settable=True,
data='Rectangular',
units=None
),
'field_master': Attr(
name='field_master',
type='LOGIC',
settable=True,
data=False,
units=None
)
}
)
Twiss parameters¶
In [8]:
Copied!
ele.twiss
ele.twiss
Out[8]:
ElementTwiss(
alpha_a=-14.5306497921953,
alpha_b=25.8281436293631,
beta_a=15.9569604975133,
beta_b=76.3766224115966,
dalpha_dpz_a=-330.217881814612,
dalpha_dpz_b=465.372281067964,
dbeta_dpz_a=366.293827131603,
dbeta_dpz_b=1607.40327209197,
deta_dpz_a=-1.31078689730276,
deta_dpz_b=0.077542150026579,
deta_dpz_x=-1.33623745152148,
deta_dpz_y=0.126869936796069,
deta_ds_a=-0.121720793069142,
deta_ds_b=-4.09604188903925e-05,
deta_dsx=-0.124087902699568,
deta_dsy=-0.0112767096416127,
detap_dpz_a=-0.703120021002026,
detap_dpz_b=-0.0356480151500083,
detap_dpz_x=-0.709366488383933,
detap_dpz_y=-0.183508814783342,
eta_a=-0.125955667075685,
eta_b=-0.00296751650073793,
eta_x=-0.128166638603888,
eta_y=-0.0126152324800141,
etap_a=-0.114114719081798,
etap_b=-4.40445098414486e-05,
etap_x=-0.116363892194681,
etap_y=-0.0106308883830499,
gamma_a=13.2944982483653,
gamma_b=8.74734941457128,
mode_flip=False,
phi_a=1.04514471916855,
phi_b=1.55860768491163
)
Orbit¶
In [9]:
Copied!
ele.orbit
ele.orbit
Out[9]:
ElementOrbit(
beta=-1.0,
charge=0.0,
direction=1,
dt_ref=0.0,
field=[0.0, 0.0],
ix_ele=-1,
location='Upstream_End',
p0c=0.0,
phase=[0.0, 0.0],
px=0.0,
py=0.0,
pz=0.0,
s=0.0,
species='Not_Set!',
spin=[0.0, 0.0, 0.0],
state='Not_Set',
t=0.0,
x=0.0,
y=0.0,
z=0.0
)
Floor coordinates¶
In [10]:
Copied!
ele.floor
ele.floor
Out[10]:
ElementFloorAll(
which='model',
beginning=ElementFloor(
which='model',
where='beginning',
actual=ElementFloorPosition(
x=0.0,
y=0.0,
z=0.637956,
theta=0.0,
phi=0.0,
psi=0.0754952,
wmat=array([[ 0.99715159, -0.07542351, 0. ],
[ 0.07542351, 0.99715159, 0. ],
[ 0. , 0. , 1. ]])
),
reference=ElementFloorPosition(
x=0.0,
y=0.0,
z=0.637956,
theta=0.0,
phi=0.0,
psi=0.0,
wmat=array([[ 1., -0., 0.],
[ 0., 1., 0.],
[-0., 0., 1.]])
),
slaves={}
),
center=ElementFloor(
which='model',
where='center',
actual=ElementFloorPosition(
x=0.0,
y=0.0,
z=1.40035600000005,
theta=0.0,
phi=0.0,
psi=0.0754952,
wmat=array([[ 0.99715159, -0.07542351, 0. ],
[ 0.07542351, 0.99715159, 0. ],
[ 0. , 0. , 1. ]])
),
reference=ElementFloorPosition(
x=0.0,
y=0.0,
z=1.40035600000005,
theta=0.0,
phi=0.0,
psi=0.0,
wmat=array([[ 1., -0., 0.],
[ 0., 1., 0.],
[-0., 0., 1.]])
),
slaves={}
),
end=ElementFloor(
which='model',
where='end',
actual=ElementFloorPosition(
x=0.0,
y=0.0,
z=2.16275600000009,
theta=0.0,
phi=0.0,
psi=0.0754952,
wmat=array([[ 0.99715159, -0.07542351, 0. ],
[ 0.07542351, 0.99715159, 0. ],
[ 0. , 0. , 1. ]])
),
reference=ElementFloorPosition(
x=0.0,
y=0.0,
z=2.16275600000009,
theta=0.0,
phi=0.0,
psi=0.0,
wmat=array([[ 1., -0., 0.],
[ 0., 1., 0.],
[-0., 0., 1.]])
),
slaves={}
)
)
Transfer matrix (mat6)¶
In [11]:
Copied!
ele.mat6
ele.mat6
Out[11]:
ElementMat6(
which='model',
vec0=array([-5.96034007e-08, -2.80046616e-08, -7.05420754e-09, -2.13217042e-09,
1.64134376e-05, 0.00000000e+00]),
mat6=array([[ 2.13280675e+00, 2.06565046e+00, 1.06264456e-01,
-5.19203611e-03, 0.00000000e+00, -8.32058018e-03],
[ 1.72825682e+00, 2.13876037e+00, 1.85340334e-01,
7.01049607e-02, 0.00000000e+00, -3.90942512e-03],
[ 2.00805572e-01, 1.40422386e-01, 1.80785561e-01,
1.07619865e+00, 0.00000000e+00, -9.84760915e-04],
[ 1.80201468e-01, 1.67905099e-01, -8.91733320e-01,
1.76484502e-01, 0.00000000e+00, -2.97649038e-04],
[-6.15973697e-03, -9.84377110e-03, -1.94751404e-04,
-4.57077370e-04, 1.00000000e+00, 4.28675359e-05],
[ 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,
0.00000000e+00, 0.00000000e+00, 1.00000000e+00]]),
symplectic_error=5.10702591327572e-15
)
Getting multiple Elements¶
Use tao.eles() to query multiple elements at once. It supports names, indices, ranges, wildcards, and key-based matching.
In [12]:
Copied!
# By index range
elements = tao.eles("1:5")
for ele in elements:
print(f"{ele.name:20s} key={ele.key:15s} s={ele.head.s:.3f}")
# By index range
elements = tao.eles("1:5")
for ele in elements:
print(f"{ele.name:20s} key={ele.key:15s} s={ele.head.s:.3f}")
IP_L0 key=Marker s=0.000 CLEO_SOL#3 key=Solenoid s=0.622 DET_00W key=Marker s=0.622 CLEO_SOL#4 key=Solenoid s=0.638 Q00W\CLEO_SOL key=Sol_Quad s=1.755
In [13]:
Copied!
# By wildcard
quads = tao.eles("Q0*")
for q in quads:
print(f"{q.name:20s} K1={q.attrs['K1'].data}")
# By wildcard
quads = tao.eles("Q0*")
for q in quads:
print(f"{q.name:20s} K1={q.attrs['K1'].data}")
Q00W K1=-0.8417848364530158 Q00W\CLEO_SOL K1=-0.8417848364530158 Q00W#1 K1=-0.8417848364530158 Q01W K1=0.639155 Q02W K1=-0.206328 Q03W K1=-0.128947 Q04W K1=0.420143 Q05W K1=-0.373608 Q06W K1=0.421931 Q07W K1=-0.246488 Q08W K1=0.144685 Q09W K1=-0.19165 Q09E K1=-0.190367 Q08E K1=0.144685 Q07E K1=-0.246488 Q06E K1=0.421931 Q05E K1=-0.373608 Q04E K1=0.420143 Q03E K1=-0.128947 Q02E K1=-0.206328 Q01E K1=0.639155 Q00E K1=-0.8417848364530158 Q00E#1 K1=-0.8417848364530158 Q00E\CLEO_SOL K1=-0.8417848364530158
In [14]:
Copied!
# By element key - get all quadrupoles
all_quads = tao.eles("quad::*")
print(f"Found {len(all_quads)} quadrupoles")
# By element key - get all quadrupoles
all_quads = tao.eles("quad::*")
print(f"Found {len(all_quads)} quadrupoles")
Found 126 quadrupoles
Controlling what data is loaded¶
Loading all data for many elements can be slow. You can control which fields are loaded.
In [15]:
Copied!
# Minimal element - only head data
ele_minimal = tao.ele("Q00W", defaults=False)
print(f"twiss: {ele_minimal.twiss}")
print(f"orbit: {ele_minimal.orbit}")
# Minimal element - only head data
ele_minimal = tao.ele("Q00W", defaults=False)
print(f"twiss: {ele_minimal.twiss}")
print(f"orbit: {ele_minimal.orbit}")
twiss: None orbit: None
In [16]:
Copied!
# Only specific fields
ele_twiss_only = tao.ele("Q00W", defaults=False, twiss=True, attrs=True)
print(f"twiss: {ele_twiss_only.twiss is not None}")
print(f"orbit: {ele_twiss_only.orbit is not None}")
# Only specific fields
ele_twiss_only = tao.ele("Q00W", defaults=False, twiss=True, attrs=True)
print(f"twiss: {ele_twiss_only.twiss is not None}")
print(f"orbit: {ele_twiss_only.orbit is not None}")
twiss: True orbit: False
In [17]:
Copied!
# Defaults on, but exclude orbit
ele_no_orbit = tao.ele("Q00W", orbit=False)
print(f"twiss: {ele_no_orbit.twiss is not None}")
print(f"orbit: {ele_no_orbit.orbit is not None}")
# Defaults on, but exclude orbit
ele_no_orbit = tao.ele("Q00W", orbit=False)
print(f"twiss: {ele_no_orbit.twiss is not None}")
print(f"orbit: {ele_no_orbit.orbit is not None}")
twiss: True orbit: False
You can also change the global defaults for all future queries:
In [18]:
Copied!
from pytao.model import Element
# See current defaults
print("Current defaults:", Element.DEFAULTS)
from pytao.model import Element
# See current defaults
print("Current defaults:", Element.DEFAULTS)
Current defaults: {'bunch_params', 'wall3d', 'lord_slave', 'chamber_walls', 'grid_field', 'control_vars', 'multipoles', 'wake', 'attrs', 'floor', 'twiss', 'photon', 'orbit', 'mat6'}
Lattice objects¶
A Lattice is a collection of Elements. There are several constructors depending on what subset of the lattice you need.
In [19]:
Copied!
from pytao.model import Lattice
# Load tracking elements (skips lords/markers)
lat = Lattice.from_tao_tracking(tao)
print(lat)
from pytao.model import Lattice
# Load tracking elements (skips lords/markers)
lat = Lattice.from_tao_tracking(tao)
print(lat)
<Lattice which=model with 870 elements>
In [20]:
Copied!
# Load unique (non-slave) elements
lat_unique = Lattice.from_tao_unique(tao)
print(lat_unique)
# Load unique (non-slave) elements
lat_unique = Lattice.from_tao_unique(tao)
print(lat_unique)
<Lattice which=model with 1010 elements>
In [21]:
Copied!
# Load a subset by range
lat_subset = Lattice.from_tao_tracking(tao, track_start="1", track_end="20")
print(lat_subset)
# Load a subset by range
lat_subset = Lattice.from_tao_tracking(tao, track_start="1", track_end="20")
print(lat_subset)
<Lattice which=model with 20 elements>
Accessing elements in a Lattice¶
Lattice objects provide convenient dictionaries for lookup.
In [22]:
Copied!
# By name
by_name = lat.by_element_name
list(by_name.keys())[:10]
# By name
by_name = lat.by_element_name
list(by_name.keys())[:10]
Out[22]:
['BEGINNING', 'IP_L0', 'CLEO_SOL#3', 'DET_00W', 'CLEO_SOL#4', 'Q00W\\CLEO_SOL', 'Q00W#1', 'D003', 'DET_01W', 'D004']
In [23]:
Copied!
# By element type (key)
by_key = lat.by_element_key
print("Element types:", list(by_key.keys()))
# By element type (key)
by_key = lat.by_element_key
print("Element types:", list(by_key.keys()))
Element types: ['Beginning_Ele', 'Marker', 'Solenoid', 'Sol_Quad', 'Quadrupole', 'Drift', 'Kicker', 'SBend', 'Wiggler', 'ELSeparator', 'Sextupole', 'RFCavity', 'Octupole']
In [24]:
Copied!
# By index
by_index = lat.by_element_index
by_index[1].name
# By index
by_index = lat.by_element_index
by_index[1].name
Out[24]:
'IP_L0'
Plotting with Element data¶
Since all data is in Python objects, it's easy to plot with matplotlib or any other library.
In [25]:
Copied!
import matplotlib.pyplot as plt
import numpy as np
import matplotlib.pyplot as plt
import numpy as np
In [26]:
Copied!
# Extract Twiss data from lattice elements
eles = lat.elements
s_values = [ele.head.s for ele in eles]
alpha_a = [ele.twiss.alpha_a for ele in eles]
alpha_b = [ele.twiss.alpha_b for ele in eles]
fig, ax = plt.subplots(figsize=(12, 4))
ax.plot(s_values, alpha_a, label=r"$\alpha_a$")
ax.plot(s_values, alpha_b, label=r"$\alpha_b$")
ax.set_xlabel("s (m)")
ax.set_ylabel(r"$\alpha_A$, $\alpha_B$ (m)")
ax.set_title("Twiss Alpha Functions")
ax.legend()
plt.tight_layout()
# Extract Twiss data from lattice elements
eles = lat.elements
s_values = [ele.head.s for ele in eles]
alpha_a = [ele.twiss.alpha_a for ele in eles]
alpha_b = [ele.twiss.alpha_b for ele in eles]
fig, ax = plt.subplots(figsize=(12, 4))
ax.plot(s_values, alpha_a, label=r"$\alpha_a$")
ax.plot(s_values, alpha_b, label=r"$\alpha_b$")
ax.set_xlabel("s (m)")
ax.set_ylabel(r"$\alpha_A$, $\alpha_B$ (m)")
ax.set_title("Twiss Alpha Functions")
ax.legend()
plt.tight_layout()
In [27]:
Copied!
tao.plot("alpha", width=12)
tao.plot("alpha", width=12)
Saving and loading¶
Element and Lattice objects can be serialized to JSON, compressed JSON, or msgpack (the fastest!).
In [28]:
Copied!
from pathlib import Path
lat_subset = Lattice.from_tao_tracking(tao, track_start="1", track_end="20")
# Save to JSON
lat_subset.write(Path("lattice_subset.json"))
# Save to compressed JSON
lat_subset.write(Path("lattice_subset.json.gz"))
# Save to msgpack
lat_subset.write(Path("lattice_subset.msgpack"));
# Note that the above returns all of the data that was just written to file.
# The semicolon at the end suppresses this output in Jupyter notebooks.
from pathlib import Path
lat_subset = Lattice.from_tao_tracking(tao, track_start="1", track_end="20")
# Save to JSON
lat_subset.write(Path("lattice_subset.json"))
# Save to compressed JSON
lat_subset.write(Path("lattice_subset.json.gz"))
# Save to msgpack
lat_subset.write(Path("lattice_subset.msgpack"));
# Note that the above returns all of the data that was just written to file.
# The semicolon at the end suppresses this output in Jupyter notebooks.
In [29]:
Copied!
# Load back from any format
for path in ["lattice_subset.json", "lattice_subset.json.gz", "lattice_subset.msgpack"]:
print("Loading:", path)
lat_loaded = Lattice.from_file(path)
print(lat_loaded)
print("Equal to original?", lat_loaded == lat_subset)
print()
# Load back from any format
for path in ["lattice_subset.json", "lattice_subset.json.gz", "lattice_subset.msgpack"]:
print("Loading:", path)
lat_loaded = Lattice.from_file(path)
print(lat_loaded)
print("Equal to original?", lat_loaded == lat_subset)
print()
Loading: lattice_subset.json <Lattice file=lattice_subset.json which=model with 20 elements> Equal to original? True Loading: lattice_subset.json.gz <Lattice file=lattice_subset.json.gz which=model with 20 elements> Equal to original? True Loading: lattice_subset.msgpack <Lattice file=lattice_subset.msgpack which=model with 20 elements>
Equal to original? True
In [30]:
Copied!
# Cleanup
for fn in ["lattice_subset.json", "lattice_subset.json.gz", "lattice_subset.msgpack"]:
fn = Path(fn)
if fn.exists():
fn.unlink()
# Cleanup
for fn in ["lattice_subset.json", "lattice_subset.json.gz", "lattice_subset.msgpack"]:
fn = Path(fn)
if fn.exists():
fn.unlink()