quimb.experimental.tnvmc¶
Tools for generic VMC optimization of tensor networks.
Module Contents¶
Classes¶
A convenience context manager that does nothing. |
|
Keep track of the windowed mean and estimated variance of a stream of |
|
Sampler that explicitly constructs the full probability distribution. |
|
|
|
Functions¶
|
Return a copy of |
|
|
|
|
|
|
|
|
|
Call the function |
|
|
Attributes¶
- quimb.experimental.tnvmc.sample_bitstring_from_prob_ndarray(p, rng)¶
- quimb.experimental.tnvmc.shuffled(it)¶
Return a copy of
itin random order.
- class quimb.experimental.tnvmc.NoContext¶
A convenience context manager that does nothing.
- __enter__()¶
- __exit__(*_, **__)¶
- class quimb.experimental.tnvmc.MovingStatistics(window_size)¶
Keep track of the windowed mean and estimated variance of a stream of values on the fly.
- property mean¶
- property var¶
- property std¶
- property err¶
- update(x)¶
- class quimb.experimental.tnvmc.DenseSampler(psi=None, seed=None, **contract_opts)¶
Sampler that explicitly constructs the full probability distribution. Useful for debugging small problems.
- _set_psi(psi)¶
- sample()¶
- update(**kwargs)¶
- class quimb.experimental.tnvmc.DirectTNSampler(tn, sweeps=1, max_group_size=8, chi=None, optimize=None, optimize_share_path=False, seed=None, track=False)¶
- Parameters:
tn (TensorNetwork) – The tensor network to sample from.
sweeps (int, optional) – The number of sweeps to perform.
max_group_size (int, optional) – The maximum number of sites to include in a single marginal.
chi (int, optional) – The maximum bond dimension to use for compressed contraction.
optimize (PathOptimizer, optional) – The path optimizer to use.
optimize_share_path (bool, optional) – If
True, a single path will be used for all contractions regardless of which marginal (i.e. which indices are open) is begin computed.
- plot()¶
- calc_groups(**kwargs)¶
Calculate how to group the sites into marginals.
- get_groups()¶
- calc_path()¶
- get_path()¶
- get_optimize()¶
- contract(tn, output_inds)¶
- sample()¶
- quimb.experimental.tnvmc.compute_amplitude(tn, config, chi, optimize)¶
- quimb.experimental.tnvmc.compute_amplitudes(tn, configs, chi, optimize)¶
- quimb.experimental.tnvmc.compute_local_energy(ham, tn, config, chi, optimize)¶
- quimb.experimental.tnvmc.draw_config(edges, config)¶
- class quimb.experimental.tnvmc.ClusterSampler(psi=None, max_distance=1, use_gauges=True, seed=None, contract_opts=None)¶
- candidate¶
- _set_psi(psi)¶
- sample()¶
- accept(config)¶
- update(**kwargs)¶
- class quimb.experimental.tnvmc.ExchangeSampler(edges, seed=None)¶
- candidate()¶
- accept(config)¶
- sample()¶
- update(**_)¶
- class quimb.experimental.tnvmc.HamiltonianSampler(ham, seed=None)¶
- candidate()¶
- accept(config)¶
- sample()¶
- update(**_)¶
- class quimb.experimental.tnvmc.MetropolisHastingsSampler(sub_sampler, amplitude_factory=None, initial=None, burn_in=0, seed=None, track=False)¶
- property acceptance_ratio¶
- sample()¶
- update(**kwargs)¶
- plot()¶
Call the function
func, which should be an array function making use of autoray dispatched calls, multiple times, automatically reusing shared intermediates.
- class quimb.experimental.tnvmc.ComposePartial(f, f_args, f_kwargs, g)¶
- __slots__ = ('f', 'f_args', 'f_kwargs', 'g')¶
- __call__(*args, **kwargs)¶
- quimb.experimental.tnvmc._partial_compose_cache¶
- quimb.experimental.tnvmc.get_compose_partial(f, f_args, f_kwargs, g)¶
- quimb.experimental.tnvmc.fuse_unary_ops_(Z)¶
- class quimb.experimental.tnvmc.AmplitudeFactory(psi=None, contract_fn=None, maxsize=2**20, autojit_opts=(), **contract_opts)¶
- _set_psi(psi)¶
- compute_single(config)¶
Compute the amplitude of
config, making use of autojit.
- compute_multi(configs)¶
Compute the amplitudes corresponding to the sequence
configs, making use of shared intermediates.
- amplitude(config)¶
Get the amplitude of
config, either from the cache or by computing it.
- amplitudes(configs)¶
- prob(config)¶
Calculate the probability of a configuration.
- clear()¶
- __contains__(config)¶
- __setitem__(config, c)¶
- __getitem__(config)¶
- __repr__()¶
Return repr(self).
- class quimb.experimental.tnvmc.GradientAccumulator¶
- _init_storage(grads)¶
- update(grads_logpsi_sample, local_energy)¶
- extract_grads_energy()¶
- class quimb.experimental.tnvmc.SGD(learning_rate=0.01)¶
Bases:
GradientAccumulator- transform_gradients()¶
- class quimb.experimental.tnvmc.SignDescent(learning_rate=0.01)¶
Bases:
GradientAccumulator- transform_gradients()¶
- class quimb.experimental.tnvmc.RandomSign(learning_rate=0.01)¶
Bases:
GradientAccumulator- transform_gradients()¶
- class quimb.experimental.tnvmc.Adam(learning_rate=0.01, beta1=0.9, beta2=0.999, eps=1e-08)¶
Bases:
GradientAccumulator- transform_gradients()¶
- class quimb.experimental.tnvmc.StochasticReconfigureGradients(delta=1e-05)¶
- update(grads_logpsi_sample, local_energy)¶
- extract_grads_energy()¶
- class quimb.experimental.tnvmc.SR(learning_rate=0.05, delta=1e-05)¶
Bases:
SGD,StochasticReconfigureGradients
- class quimb.experimental.tnvmc.SRADAM(learning_rate=0.01, beta1=0.9, beta2=0.999, eps=1e-08, delta=1e-05)¶
Bases:
Adam,StochasticReconfigureGradients
- class quimb.experimental.tnvmc.TNVMC(psi, ham, sampler, conditioner='auto', learning_rate=0.01, optimizer='adam', optimizer_opts=None, track_window_size=1000, **contract_opts)¶
- _compute_log_gradients_torch(config)¶
- _compute_local_energy(config)¶
- _run(steps, batchsize)¶
- run(total=10000, batchsize=100, progbar=True)¶
- measure(max_samples=10000, rtol=0.0001, progbar=True)¶
- plot(figsize=(12, 6), yrange_quantile=(0.01, 0.99), zoom='auto', hlines=())¶