Sample independent qubit sets without merging state space #4110

daxfohl · 2021-05-15T21:44:55Z

Next step of #3409 (first step was #4100)

#4100 was a big change in that it allows independent state spaces to evolve independently, exponentially reducing the runtime of simulation when states are not entangled. However we still generate the fully-merged state space each time we create a StepResult, so the effect of those improvements is limited. This PR fixes that.

This PR avoids merging the whole state space after each moment to create the StepResult. Instead, the simulators pass the raw OperationTarget (the lookup of unentangled states) into the StepResult (StepResultBase created to handle common functionality). That only merges the full state when the _simulator_state() function is explicitly called. In particular, sampling from the StepResults is done by independently sampling the substates and zipping the results together, avoiding creating the fully merged state tensor.

The final PR will be #4198, which will also handle this for TrialResult, and thus allow end to end simulation of very large unentangled circuits without running out of memory.

BREAKING CHANGES:

The behavior in test_state_vector_copy and test_density_matrix_copy, where if you do a step_result.get_state(copy=False) and manually fiddle with that vector, and the changes to that vector are preserved, no longer works by default. However this behavior is not documented, and seems unusual to expect. If someone requires the behavior, it will continue to work if split_untangled_qubits=False, but this PR sets split_untangled_qubits=True by default.
~~Sampling is not deterministic even with same seed. ActOnArgsContainer.sample iterates a set, which appears to be non-deterministic ordering in Python. I could probably figure out a way around this.~~
Rounding errors may be affected. From observation, setting split_untangled_qubits=True seems to have a slightly worse effect on rounding errors and ATOLs may be affected. (Or actually it's likely the rounding error is equal magnitude but just different--I only notice the tests where tolerances need adjusted downward, not vice-versa).

95-martin-orion · 2021-07-08T13:18:10Z

@daxfohl, is this PR now up-to-date with master such that changes from #4100 do not also appear in the "changed files" here?

daxfohl · 2021-07-08T15:18:31Z

is this PR now up-to-date with master such that changes from #4100 do not also appear in the "changed files" here?

Yes.

95-martin-orion

Full review pass. The issue I'm most concerned about here (and that is most likely to override other issues listed below) is the step_result._sim_state mutation concern.

cirq-core/cirq/sim/simulator_base.py

cirq-core/cirq/sim/simulator_base_test.py

examples/bb84.py

cirq-core/cirq/sim/density_matrix_simulator.py

cirq-core/cirq/sim/act_on_args_container.py

daxfohl

I can't seem to fully cancel this review.

CirqBot · 2021-07-12T20:17:29Z

Automerge cancelled: A required status check is not present.

Missing statuses: ['cla/google']

…#4110) * split * Allow config param split_entangled_states * default split to off * ensure consistent_act_on circuits have a qubit. * lint * lint * mps * lint * lint * run sparse by default * fix tests * fix tests * fix tests * most of sparse and dm * clifford * sim_base * sim_base * mps * turn off on experiments with rounding error * fix tests * fix tests * fix testsCreate base step result * clifford * mps * mps * mps * tableau * test simulator * test simulator * Update simulator_base.py * Drop mps/join * Fix clifford extract * lint * simplify index * Add qubits to base class * Fix clifford sampling * Fix _sim_state_values * fix tostring tests, format * remove split/join from ch-form * remove split/join from ch-form * push merged state to base layer * lint * mypy * mypy * mypy * Add default arg for zero qubit circuits * Have last repetition reuse original state repr * Remove cast * Split all pure initial states by default * Detangle on reset channels * docstrings * docstrings * docstrings * docstrings * fix merge * lint * Add unit test for integer states * format * Add tests for splitting and joining * remove unnecessary qubits param * Clean up default args * Fix failing test * Add ActOnArgsContainer * Add ActOnArgsContainer * Clean up tests * Clean up tests * Clean up tests * format * Fix tests and coverage * Add OperationTarget interface * Fix unit tests * mypy, lint, mocks, coverage * coverage * lint, tests * lint, tests * mypy * mypy, tests * remove test code * test * dead code * mocks * add log to container * fix logs * dead code * unit test * unit test * dead code * operationtarget samples * StepResultBase * Mock, format * EmptyActOnArgs * EmptyActOnArgs * simplify dummyargs * lint * Add [] to actonargs * rename _create_act_on_arg * coverage * coverage * Default sparse sim to split=false * format * Default sparse sim to split=false * Default density matrix sim to split=false * lint * lint * lint * lint * address review comments * lint * Defaults back to split=false * add error if setting state when split is enabled * Unit tests * coverage * coverage * coverage * docs * conflicts * conflicts * cover * Add qubits to bb84 * mergedsimstate private * q_set * default to split=True * Allow set_state * Allow set_state * format * fix merge * fix merge * maintain order in sampling for determinicity. * Pydoc fixes * revert bb48 num_qubits change * fix docstrings for set_state error * Remove duplicate sample declaration from ActOnArgs * Remove unnecessary split_untangled_states=True * Reduce atol of dm/sv test * Add test for sim_state propagation from step_result * Add test for sim_state propagation from step_result Co-authored-by: Cirq Bot <[email protected]>

daxfohl added 30 commits May 6, 2021 13:28

split

293374d

Allow config param split_entangled_states

4faff08

default split to off

8c202a4

ensure consistent_act_on circuits have a qubit.

54f1eea

lint

265d939

lint

099ecc5

mps

a698893

lint

251eeb4

lint

a255905

run sparse by default

6f58842

fix tests

b7254f9

fix tests

79e77b2

fix tests

4d301e5

most of sparse and dm

181d5fb

clifford

f982731

sim_base

1ee3468

sim_base

994d9d4

mps

fe2761c

turn off on experiments with rounding error

3c472b6

fix tests

c4bd708

fix tests

1ed1f2e

fix testsCreate base step result

d0cd4f2

clifford

ab1f63c

mps

bedb9ec

mps

8b34b11

mps

58ed8fa

Merge branch 'master' into split

ba93627

Merge branch 'split' into sample

027e1b6

Merge branch 'sample' into sample2

f44b863

tableau

cba2940

95-martin-orion self-requested a review July 8, 2021 13:14

95-martin-orion self-assigned this Jul 8, 2021

95-martin-orion requested changes Jul 8, 2021

View reviewed changes

daxfohl added 6 commits July 8, 2021 18:24

Pydoc fixes

1111cf0

revert bb48 num_qubits change

bd43695

fix docstrings for set_state error

802958f

Merge branch 'master' into sample

da06988

Remove duplicate sample declaration from ActOnArgs

c9a41d7

Remove unnecessary split_untangled_states=True

6e2cdff

daxfohl commented Jul 9, 2021

View reviewed changes

daxfohl added 3 commits July 8, 2021 21:20

Reduce atol of dm/sv test

365e4bd

Add test for sim_state propagation from step_result

d90e2dc

Add test for sim_state propagation from step_result

bd2bf74

95-martin-orion added the BREAKING CHANGE For pull requests that are important to mention in release notes. label Jul 12, 2021

95-martin-orion approved these changes Jul 12, 2021

View reviewed changes

95-martin-orion added the automerge Tells CirqBot to sync and merge this PR. (If it's running.) label Jul 12, 2021

CirqBot added the front_of_queue_automerge CirqBot uses this label to indicate (and remember) what's being merged next. label Jul 12, 2021

Merge branch 'master' into sample

b28f2aa

CirqBot removed automerge Tells CirqBot to sync and merge this PR. (If it's running.) front_of_queue_automerge CirqBot uses this label to indicate (and remember) what's being merged next. labels Jul 12, 2021

95-martin-orion merged commit c28646f into quantumlib:master Jul 12, 2021

daxfohl deleted the sample branch July 13, 2021 13:34

daxfohl restored the sample branch July 13, 2021 13:37

daxfohl deleted the sample branch March 30, 2022 14:20

daxfohl mentioned this pull request Apr 2, 2022

Remove some 0.15 items from cirq.sim #5137

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Sample independent qubit sets without merging state space #4110

Sample independent qubit sets without merging state space #4110

daxfohl commented May 15, 2021 •

edited

Loading

95-martin-orion commented Jul 8, 2021

daxfohl commented Jul 8, 2021

95-martin-orion left a comment

daxfohl left a comment

CirqBot commented Jul 12, 2021

Sample independent qubit sets without merging state space #4110

Sample independent qubit sets without merging state space #4110

Conversation

daxfohl commented May 15, 2021 • edited Loading

95-martin-orion commented Jul 8, 2021

daxfohl commented Jul 8, 2021

95-martin-orion left a comment

Choose a reason for hiding this comment

daxfohl left a comment

Choose a reason for hiding this comment

CirqBot commented Jul 12, 2021

daxfohl commented May 15, 2021 •

edited

Loading