Rename batch to graph #216
Conversation
|
""" WalkthroughThe changes systematically rename variables, function names, comments, and documentation from "batch" to "graph" throughout the codebase, reflecting a shift in terminology to better represent the data structure. This includes updates in scripts, core modules, tests, and documentation, without altering underlying logic or control flow. Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant Script/Module
participant State
participant Integrator
User->>Script/Module: Run computation or test
Script/Module->>State: Access n_atoms_per_graph
State-->>Script/Module: Return per-graph atom counts
Script/Module->>Integrator: Pass per-graph data
Integrator-->>Script/Module: Process and return results
Script/Module-->>User: Output results with per-graph terminology
Possibly related PRs
Suggested labels
Poem
📜 Recent review detailsConfiguration used: CodeRabbit UI 📒 Files selected for processing (2)
🚧 Files skipped from review as they are similar to previous changes (2)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (37)
✨ Finishing Touches
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
CodeRabbit Configuration File (
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
🧹 Nitpick comments (2)
examples/tutorials/state_tutorial.py (1)
263-263: Label still says “Per-batch” – update to match the new terminologyThe printout now queries
scope["per_graph"], but the user-facing label remains “Per-batch”. Tiny mismatch, but it’s user-visible.-print("Per-batch properties:", scope["per_graph"]) +print("Per-graph properties:", scope["per_graph"])tests/test_integrators.py (1)
375-409: Docstring/comments lag behind rename + minor ratio comment
- The test has been renamed to
atoms_per_graph, yet the docstring still says “…per batch”.- Inline comment line 382 says “8:1 atom ratio” but the code creates 8 vs 64 atoms (1 : 8); could confuse future readers.
-"""Test that compute_cell_force correctly scales by number of atoms per batch. +"""Test that _compute_cell_force correctly scales by the number of atoms per graph. @@ - # Setup minimal state with two batches having 8:1 atom ratio + # Two graphs with an 8 × larger atom count in the second (8 vs 64)
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (12)
.gitignore(1 hunks)examples/scripts/1_Introduction/1.2_MACE.py(1 hunks)examples/scripts/2_Structural_optimization/2.3_MACE_Gradient_Descent.py(1 hunks)examples/tutorials/state_tutorial.py(1 hunks)tests/test_integrators.py(1 hunks)tests/test_state.py(3 hunks)tests/test_trajectory.py(1 hunks)torch_sim/integrators/npt.py(10 hunks)torch_sim/integrators/nvt.py(2 hunks)torch_sim/io.py(3 hunks)torch_sim/quantities.py(1 hunks)torch_sim/state.py(14 hunks)
🧰 Additional context used
🧬 Code Graph Analysis (3)
torch_sim/integrators/nvt.py (1)
torch_sim/state.py (1)
n_atoms_per_graph(152-158)
tests/test_state.py (1)
torch_sim/state.py (1)
n_atoms_per_graph(152-158)
tests/test_trajectory.py (2)
tests/conftest.py (1)
si_double_sim_state(323-325)torch_sim/trajectory.py (1)
get_array(601-629)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (58)
- GitHub Check: test-examples (examples/scripts/2_Structural_optimization/2.6_MACE_UnitCellFilter_FIRE.py)
- GitHub Check: test-examples (examples/scripts/4_High_level_api/4.1_high_level_api.py)
- GitHub Check: test-examples (examples/scripts/5_Workflow/5.3_Elastic.py)
- GitHub Check: test-examples (examples/scripts/2_Structural_optimization/2.5_MACE_UnitCellFilter_Gradient_Descen...
- GitHub Check: test-examples (examples/scripts/4_High_level_api/4.2_auto_batching_api.py)
- GitHub Check: test-examples (examples/scripts/5_Workflow/5.1_a2c_silicon_batched.py)
- GitHub Check: test-examples (examples/scripts/2_Structural_optimization/2.3_MACE_Gradient_Descent.py)
- GitHub Check: test-examples (examples/scripts/2_Structural_optimization/2.1_Lennard_Jones_FIRE.py)
- GitHub Check: test-examples (examples/scripts/5_Workflow/5.2_In_Flight_WBM.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.9_MACE_NVT_staggered_stress.py)
- GitHub Check: test-examples (examples/scripts/1_Introduction/1.3_Fairchem.py)
- GitHub Check: test-examples (examples/scripts/2_Structural_optimization/2.7_MACE_FrechetCellFilter_FIRE.py)
- GitHub Check: test-examples (examples/scripts/6_Phonons/6.1_Phonons_MACE.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.10_Hybrid_swap_mc.py)
- GitHub Check: test-examples (examples/scripts/6_Phonons/6.3_Conductivity_MACE.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.12_MACE_NPT_Langevin.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.4_MACE_NVT_Langevin.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.8_MACE_NPT_Nose_Hoover.py)
- GitHub Check: test-examples (examples/scripts/1_Introduction/1.2_MACE.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.6_MACE_NVT_Nose_Hoover_temp_profile.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.3_MACE_NVE_cueq.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.5_MACE_NVT_Nose_Hoover.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.13_MACE_NVE_non_pbc.py)
- GitHub Check: test-examples (examples/scripts/7_Others/7.6_Compare_ASE_to_VV_FIRE.py)
- GitHub Check: test-examples (examples/scripts/7_Others/7.4_Velocity_AutoCorrelation.py)
- GitHub Check: test-examples (examples/scripts/6_Phonons/6.2_QuasiHarmonic_MACE.py)
- GitHub Check: test-examples (examples/scripts/3_Dynamics/3.2_MACE_NVE.py)
- GitHub Check: test-examples (examples/tutorials/metatomic_tutorial.py)
- GitHub Check: test-examples (examples/tutorials/high_level_tutorial.py)
- GitHub Check: test-examples (examples/tutorials/hybrid_swap_tutorial.py)
- GitHub Check: test-examples (examples/tutorials/autobatching_tutorial.py)
- GitHub Check: test-examples (examples/tutorials/low_level_tutorial.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, orb, tests/models/test_orb.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, sevenn, tests/models/test_sevennet.py)
- GitHub Check: test-model (macos-14, 3.11, highest, metatomic, tests/models/test_metatomic.py)
- GitHub Check: test-model (macos-14, 3.11, highest, orb, tests/models/test_orb.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, graphpes, tests/models/test_graphpes.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, mace, tests/test_optimizers_vs_ase.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, fairchem, tests/models/test_fairchem.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, metatomic, tests/models/test_metatomic.py)
- GitHub Check: test-model (macos-14, 3.11, highest, fairchem, tests/models/test_fairchem.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, mattersim, tests/models/test_mattersim.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, mace, tests/models/test_mace.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, sevenn, tests/models/test_sevennet.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, mace, tests/test_optimizers_vs_ase.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, mace, tests/test_elastic.py)
- GitHub Check: test-model (ubuntu-latest, 3.12, lowest-direct, mace, tests/models/test_mace.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, orb, tests/models/test_orb.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, graphpes, tests/models/test_graphpes.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, fairchem, tests/models/test_fairchem.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, metatomic, tests/models/test_metatomic.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, mattersim, tests/models/test_mattersim.py)
- GitHub Check: test-model (ubuntu-latest, 3.11, highest, mace, tests/test_elastic.py)
- GitHub Check: test-core (macos-14, 3.12, lowest-direct)
- GitHub Check: test-core (macos-14, 3.11, highest)
- GitHub Check: test-core (ubuntu-latest, 3.12, lowest-direct)
- GitHub Check: test-core (ubuntu-latest, 3.11, highest)
- GitHub Check: build-docs
🔇 Additional comments (11)
.gitignore (1)
38-40: Ignoring.vscode/is a sound additionPrevents IDE-specific metadata from slipping into commits and keeps the repo clean.
tests/test_trajectory.py (1)
681-684: LGTM – variable rename & assertions consistent with new terminology
No functional issues spotted.torch_sim/quantities.py (1)
63-69: LGTM! Consistent terminology update.The variable rename from
dof_per_batchtodof_per_graphaligns with the broader codebase terminology update while preserving the calculation logic.tests/test_state.py (1)
32-32: LGTM! Test assertions correctly updated.The test assertions properly reflect the updated scope keys from
"per_batch"to"per_graph".Also applies to: 53-53
torch_sim/io.py (3)
229-231: LGTM! Consistent variable renaming.The local variable rename from
atoms_per_batchtoatoms_per_graphimproves terminology consistency while preserving the batch index creation logic.
301-303: LGTM! Consistent variable renaming.The variable rename maintains the same logic for creating batch indices using
torch.repeat_interleave.
372-374: LGTM! Consistent variable renaming.The rename aligns with the broader terminology update while preserving the batch index generation logic.
torch_sim/integrators/nvt.py (2)
373-380: LGTM! Consistent terminology update in initialization.The variable renames from
n_atoms_per_batch/dof_per_batchton_atoms_per_graph/dof_per_graphpreserve the degrees of freedom calculation logic while improving terminology consistency.
482-483: LGTM! Consistent variable renaming in invariant calculation.The rename to
n_atoms_per_graphmaintains the same degrees of freedom calculation for the conserved quantity computation.torch_sim/integrators/npt.py (1)
139-139: LGTM: Systematic and consistent renaming from "batch" to "graph" terminology.The renaming of variables and properties from
n_atoms_per_batch/KE_per_batch/DOF_per_batchton_atoms_per_graph/KE_per_graph/DOF_per_graphis systematic and preserves all underlying mathematical logic. The terminology change better reflects that each "batch" represents a separate molecular graph/system.Also applies to: 666-666, 672-672, 738-739, 1020-1021, 1216-1224, 1239-1239, 1288-1289, 1428-1429, 1599-1605, 1609-1611, 1621-1628
torch_sim/state.py (1)
152-152: Breaking change: Property renamed fromn_atoms_per_batchton_atoms_per_graph.This property name change supports the terminology update throughout the codebase but constitutes a breaking change for external users.
examples/scripts/2_Structural_optimization/2.3_MACE_Gradient_Descent.py
Outdated
Show resolved
Hide resolved
There was a problem hiding this comment.
Actionable comments posted: 3
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (66)
examples/scripts/1_Introduction/1.2_MACE.py(1 hunks)examples/scripts/2_Structural_optimization/2.3_MACE_Gradient_Descent.py(1 hunks)examples/scripts/2_Structural_optimization/2.5_MACE_UnitCellFilter_Gradient_Descent.py(1 hunks)examples/scripts/2_Structural_optimization/2.6_MACE_UnitCellFilter_FIRE.py(1 hunks)examples/scripts/2_Structural_optimization/2.7_MACE_FrechetCellFilter_FIRE.py(1 hunks)examples/scripts/3_Dynamics/3.10_Hybrid_swap_mc.py(1 hunks)examples/scripts/3_Dynamics/3.11_Lennard_Jones_NPT_Langevin.py(2 hunks)examples/scripts/3_Dynamics/3.12_MACE_NPT_Langevin.py(5 hunks)examples/scripts/3_Dynamics/3.13_MACE_NVE_non_pbc.py(1 hunks)examples/scripts/3_Dynamics/3.2_MACE_NVE.py(1 hunks)examples/scripts/3_Dynamics/3.3_MACE_NVE_cueq.py(1 hunks)examples/scripts/3_Dynamics/3.4_MACE_NVT_Langevin.py(1 hunks)examples/scripts/3_Dynamics/3.5_MACE_NVT_Nose_Hoover.py(1 hunks)examples/scripts/3_Dynamics/3.6_MACE_NVT_Nose_Hoover_temp_profile.py(1 hunks)examples/scripts/3_Dynamics/3.7_Lennard_Jones_NPT_Nose_Hoover.py(2 hunks)examples/scripts/3_Dynamics/3.8_MACE_NPT_Nose_Hoover.py(4 hunks)examples/scripts/3_Dynamics/3.9_MACE_NVT_staggered_stress.py(1 hunks)examples/scripts/4_High_level_api/4.2_auto_batching_api.py(1 hunks)examples/scripts/5_Workflow/5.2_In_Flight_WBM.py(1 hunks)examples/scripts/7_Others/7.3_Batched_neighbor_list.py(2 hunks)examples/scripts/7_Others/7.6_Compare_ASE_to_VV_FIRE.py(10 hunks)examples/tutorials/autobatching_tutorial.py(2 hunks)examples/tutorials/high_level_tutorial.py(1 hunks)examples/tutorials/hybrid_swap_tutorial.py(1 hunks)examples/tutorials/low_level_tutorial.py(1 hunks)examples/tutorials/state_tutorial.py(6 hunks)tests/models/conftest.py(1 hunks)tests/test_autobatching.py(6 hunks)tests/test_correlations.py(1 hunks)tests/test_integrators.py(6 hunks)tests/test_io.py(6 hunks)tests/test_monte_carlo.py(5 hunks)tests/test_neighbors.py(3 hunks)tests/test_optimizers.py(9 hunks)tests/test_runners.py(9 hunks)tests/test_state.py(18 hunks)tests/test_trajectory.py(5 hunks)tests/test_transforms.py(10 hunks)torch_sim/autobatching.py(12 hunks)torch_sim/integrators/md.py(7 hunks)torch_sim/integrators/npt.py(51 hunks)torch_sim/integrators/nve.py(5 hunks)torch_sim/integrators/nvt.py(10 hunks)torch_sim/io.py(12 hunks)torch_sim/math.py(1 hunks)torch_sim/models/fairchem.py(1 hunks)torch_sim/models/graphpes.py(1 hunks)torch_sim/models/interface.py(4 hunks)torch_sim/models/lennard_jones.py(2 hunks)torch_sim/models/mace.py(9 hunks)torch_sim/models/metatomic.py(2 hunks)torch_sim/models/morse.py(2 hunks)torch_sim/models/orb.py(3 hunks)torch_sim/models/particle_life.py(2 hunks)torch_sim/models/sevennet.py(2 hunks)torch_sim/models/soft_sphere.py(4 hunks)torch_sim/monte_carlo.py(5 hunks)torch_sim/neighbors.py(3 hunks)torch_sim/optimizers.py(38 hunks)torch_sim/quantities.py(5 hunks)torch_sim/runners.py(8 hunks)torch_sim/state.py(23 hunks)torch_sim/trajectory.py(6 hunks)torch_sim/transforms.py(15 hunks)torch_sim/typing.py(1 hunks)torch_sim/workflows/a2c.py(1 hunks)
✅ Files skipped from review due to trivial changes (16)
- examples/scripts/2_Structural_optimization/2.6_MACE_UnitCellFilter_FIRE.py
- examples/scripts/2_Structural_optimization/2.7_MACE_FrechetCellFilter_FIRE.py
- examples/scripts/2_Structural_optimization/2.5_MACE_UnitCellFilter_Gradient_Descent.py
- examples/scripts/5_Workflow/5.2_In_Flight_WBM.py
- torch_sim/math.py
- examples/scripts/4_High_level_api/4.2_auto_batching_api.py
- examples/scripts/1_Introduction/1.2_MACE.py
- examples/tutorials/state_tutorial.py
- tests/test_transforms.py
- examples/tutorials/high_level_tutorial.py
- torch_sim/models/lennard_jones.py
- torch_sim/models/interface.py
- tests/test_neighbors.py
- torch_sim/integrators/nve.py
- torch_sim/trajectory.py
- torch_sim/transforms.py
🚧 Files skipped from review as they are similar to previous changes (7)
- tests/test_trajectory.py
- examples/scripts/2_Structural_optimization/2.3_MACE_Gradient_Descent.py
- torch_sim/io.py
- torch_sim/quantities.py
- torch_sim/integrators/nvt.py
- torch_sim/state.py
- tests/test_state.py
🧰 Additional context used
🧬 Code Graph Analysis (19)
examples/scripts/3_Dynamics/3.12_MACE_NPT_Langevin.py (1)
torch_sim/quantities.py (1)
calc_kT(23-69)
examples/tutorials/hybrid_swap_tutorial.py (1)
torch_sim/state.py (3)
n_graphs(202-204)device(142-144)dtype(147-149)
torch_sim/models/graphpes.py (1)
torch_sim/state.py (1)
n_graphs(202-204)
examples/scripts/3_Dynamics/3.10_Hybrid_swap_mc.py (1)
torch_sim/state.py (1)
n_graphs(202-204)
tests/test_optimizers.py (2)
tests/conftest.py (1)
ar_supercell_sim_state(306-310)torch_sim/state.py (2)
clone(239-255)n_graphs(202-204)
tests/test_correlations.py (1)
torch_sim/state.py (4)
n_graphs(202-204)device(142-144)dtype(147-149)split(281-290)
examples/scripts/3_Dynamics/3.5_MACE_NVT_Nose_Hoover.py (2)
torch_sim/quantities.py (1)
calc_kT(23-69)torch_sim/integrators/nvt.py (1)
nvt_nose_hoover_invariant(448-517)
examples/scripts/3_Dynamics/3.7_Lennard_Jones_NPT_Nose_Hoover.py (2)
torch_sim/quantities.py (2)
calc_kT(23-69)calc_kinetic_energy(96-130)torch_sim/integrators/npt.py (2)
momenta(70-72)npt_nose_hoover_invariant(1559-1668)
examples/tutorials/autobatching_tutorial.py (1)
torch_sim/state.py (1)
n_graphs(202-204)
tests/test_monte_carlo.py (1)
torch_sim/monte_carlo.py (1)
validate_permutation(127-141)
tests/test_runners.py (3)
torch_sim/state.py (6)
batch(175-181)batch(184-190)n_graphs(202-204)device(142-144)dtype(147-149)n_atoms(152-154)tests/conftest.py (4)
ar_double_sim_state(314-319)device(24-25)dtype(29-30)ar_supercell_sim_state(306-310)tests/test_correlations.py (1)
MockState(22-42)
torch_sim/models/soft_sphere.py (1)
torch_sim/state.py (2)
SimState(26-353)n_graphs(202-204)
torch_sim/models/particle_life.py (4)
torch_sim/models/lennard_jones.py (1)
unbatched_forward(223-354)torch_sim/models/soft_sphere.py (2)
unbatched_forward(253-378)unbatched_forward(655-803)torch_sim/models/morse.py (1)
unbatched_forward(230-344)torch_sim/state.py (1)
n_graphs(202-204)
torch_sim/runners.py (2)
torch_sim/state.py (1)
n_graphs(202-204)torch_sim/properties/correlations.py (1)
update(178-196)
torch_sim/optimizers.py (12)
torch_sim/state.py (10)
n_graphs(202-204)device(142-144)dtype(147-149)to(321-335)deform_grad(391-397)clone(239-255)DeformGradMixin(356-397)_deform_grad(379-389)row_vector_cell(226-228)row_vector_cell(231-237)torch_sim/models/interface.py (4)
device(107-109)device(112-116)dtype(119-121)dtype(124-128)tests/conftest.py (2)
device(24-25)dtype(29-30)tests/test_neighbors.py (1)
dtype(15-16)tests/models/test_graphpes.py (1)
dtype(23-25)tests/models/test_mace.py (1)
dtype(28-30)tests/models/test_sevennet.py (1)
dtype(22-24)tests/models/test_mattersim.py (1)
dtype(25-27)torch_sim/integrators/md.py (1)
velocities(44-48)torch_sim/integrators/npt.py (1)
velocities(866-872)torch_sim/integrators/nvt.py (1)
velocities(269-273)torch_sim/math.py (1)
batched_vdot(992-1022)
tests/test_integrators.py (1)
torch_sim/quantities.py (1)
calc_kT(23-69)
torch_sim/monte_carlo.py (3)
torch_sim/state.py (3)
device(142-144)n_graphs(202-204)n_atoms(152-154)tests/conftest.py (1)
device(24-25)tests/test_monte_carlo.py (1)
generator(33-36)
torch_sim/autobatching.py (1)
torch_sim/state.py (2)
n_graphs(202-204)concatenate_states(806-892)
torch_sim/integrators/npt.py (4)
torch_sim/state.py (8)
device(142-144)dtype(147-149)n_graphs(202-204)n_atoms_per_graph(157-163)volume(207-209)momenta(360-366)to(321-335)clone(239-255)torch_sim/integrators/md.py (2)
velocities(44-48)calculate_momenta(51-109)torch_sim/integrators/nvt.py (1)
velocities(269-273)torch_sim/quantities.py (1)
calc_kinetic_energy(96-130)
🪛 GitHub Check: codecov/patch
torch_sim/integrators/md.py
[warning] 83-83: torch_sim/integrators/md.py#L83
Added line #L83 was not covered by tests
[warning] 90-90: torch_sim/integrators/md.py#L90
Added line #L90 was not covered by tests
[warning] 95-96: torch_sim/integrators/md.py#L95-L96
Added lines #L95 - L96 were not covered by tests
torch_sim/autobatching.py
[warning] 1023-1023: torch_sim/autobatching.py#L1023
Added line #L1023 was not covered by tests
[warning] 1029-1029: torch_sim/autobatching.py#L1029
Added line #L1029 was not covered by tests
torch_sim/integrators/npt.py
[warning] 246-246: torch_sim/integrators/npt.py#L246
Added line #L246 was not covered by tests
[warning] 249-249: torch_sim/integrators/npt.py#L249
Added line #L249 was not covered by tests
[warning] 286-286: torch_sim/integrators/npt.py#L286
Added line #L286 was not covered by tests
[warning] 294-294: torch_sim/integrators/npt.py#L294
Added line #L294 was not covered by tests
[warning] 498-498: torch_sim/integrators/npt.py#L498
Added line #L498 was not covered by tests
[warning] 507-507: torch_sim/integrators/npt.py#L507
Added line #L507 was not covered by tests
[warning] 510-510: torch_sim/integrators/npt.py#L510
Added line #L510 was not covered by tests
[warning] 581-581: torch_sim/integrators/npt.py#L581
Added line #L581 was not covered by tests
[warning] 584-584: torch_sim/integrators/npt.py#L584
Added line #L584 was not covered by tests
[warning] 887-889: torch_sim/integrators/npt.py#L887-L889
Added lines #L887 - L889 were not covered by tests
[warning] 962-964: torch_sim/integrators/npt.py#L962-L964
Added lines #L962 - L964 were not covered by tests
[warning] 975-975: torch_sim/integrators/npt.py#L975
Added line #L975 was not covered by tests
[warning] 1013-1013: torch_sim/integrators/npt.py#L1013
Added line #L1013 was not covered by tests
[warning] 1016-1017: torch_sim/integrators/npt.py#L1016-L1017
Added lines #L1016 - L1017 were not covered by tests
[warning] 1085-1085: torch_sim/integrators/npt.py#L1085
Added line #L1085 was not covered by tests
[warning] 1150-1150: torch_sim/integrators/npt.py#L1150
Added line #L1150 was not covered by tests
[warning] 1208-1208: torch_sim/integrators/npt.py#L1208
Added line #L1208 was not covered by tests
[warning] 1212-1212: torch_sim/integrators/npt.py#L1212
Added line #L1212 was not covered by tests
[warning] 1215-1220: torch_sim/integrators/npt.py#L1215-L1220
Added lines #L1215 - L1220 were not covered by tests
[warning] 1230-1230: torch_sim/integrators/npt.py#L1230
Added line #L1230 was not covered by tests
[warning] 1269-1271: torch_sim/integrators/npt.py#L1269-L1271
Added lines #L1269 - L1271 were not covered by tests
[warning] 1284-1285: torch_sim/integrators/npt.py#L1284-L1285
Added lines #L1284 - L1285 were not covered by tests
[warning] 1409-1409: torch_sim/integrators/npt.py#L1409
Added line #L1409 was not covered by tests
[warning] 1413-1414: torch_sim/integrators/npt.py#L1413-L1414
Added lines #L1413 - L1414 were not covered by tests
[warning] 1421-1421: torch_sim/integrators/npt.py#L1421
Added line #L1421 was not covered by tests
[warning] 1424-1425: torch_sim/integrators/npt.py#L1424-L1425
Added lines #L1424 - L1425 were not covered by tests
[warning] 1434-1434: torch_sim/integrators/npt.py#L1434
Added line #L1434 was not covered by tests
[warning] 1444-1444: torch_sim/integrators/npt.py#L1444
Added line #L1444 was not covered by tests
🔇 Additional comments (101)
examples/scripts/3_Dynamics/3.13_MACE_NVE_non_pbc.py (1)
80-80: Verified:graph_idxparameter is supported end-to-end
- The
calc_kinetic_energyfunction intorch_sim/quantities.pydefinesgraph_idxas an optional parameter.- The
state.graph_idxfield is populated byatoms_to_stateintorch_sim/io.py.No further changes are needed; this renaming is correct.
examples/scripts/3_Dynamics/3.3_MACE_NVE_cueq.py (1)
74-74: Parameter renaming is consistent.The change from
batch=state.batchtograph_idx=state.graph_idxaligns with the systematic renaming throughout the codebase.examples/scripts/3_Dynamics/3.2_MACE_NVE.py (1)
91-91: Parameter renaming follows established pattern.The change from
batch=state.batchtograph_idx=state.graph_idxis consistent with the renaming pattern seen across the other files.examples/scripts/3_Dynamics/3.10_Hybrid_swap_mc.py (1)
89-89: Property renaming is correct.The change from
md_state.n_batchestomd_state.n_graphscorrectly updates the tensor size calculation to use the renamed property. Based on the code snippet fromtorch_sim/state.py, then_graphsproperty properly returns the number of distinct graphs in the system.examples/scripts/3_Dynamics/3.4_MACE_NVT_Langevin.py (1)
86-86: LGTM: Consistent parameter name update.The changes correctly update the
calc_kTfunction calls to use the newgraph_idxparameter instead ofbatch, aligning with the broader renaming effort throughout the codebase.Also applies to: 93-93
examples/scripts/3_Dynamics/3.5_MACE_NVT_Nose_Hoover.py (1)
71-71: LGTM: Consistent parameter name update.The changes correctly update the
calc_kTfunction calls to use the newgraph_idxparameter instead ofbatch, maintaining consistency with the codebase-wide renaming effort.Also applies to: 79-79
examples/scripts/3_Dynamics/3.8_MACE_NPT_Nose_Hoover.py (1)
72-72: LGTM: Consistent parameter name updates across multiple functions.The changes correctly update both
calc_kTandcalc_kinetic_energyfunction calls to use the newgraph_idxparameter instead ofbatch, maintaining consistency throughout the NPT simulation loops and final calculations.Also applies to: 89-89, 98-98, 110-110, 118-120
examples/scripts/3_Dynamics/3.11_Lennard_Jones_NPT_Langevin.py (1)
122-122: LGTM: Consistent parameter name updates.The changes correctly update both
calc_kTandcalc_kinetic_energyfunction calls to use the newgraph_idxparameter instead ofbatch, maintaining consistency with the codebase-wide renaming effort.Also applies to: 128-128, 142-142, 150-150
examples/tutorials/hybrid_swap_tutorial.py (1)
136-136: LGTM: Consistent property name update.The change correctly updates the tensor size from
md_state.n_batchestomd_state.n_graphs, aligning with the codebase-wide renaming effort. The tensor will still have the correct size for tracking permutation status per graph.examples/scripts/3_Dynamics/3.6_MACE_NVT_Nose_Hoover_temp_profile.py (1)
166-166: LGTM! Parameter name updated consistently.The change from
batch=state.batchtograph_idx=state.graph_idxcorrectly aligns with the updatedcalc_kTfunction signature and the broader semantic shift from "batch" to "graph" terminology.torch_sim/typing.py (1)
18-18: LGTM! Type definition updated to match renamed attribute.The
StateKeytype literal correctly reflects the renaming from"batch"to"graph_idx"in theSimStateclass, ensuring type safety is maintained throughout the codebase.examples/tutorials/low_level_tutorial.py (1)
231-231: LGTM! Tutorial updated with correct API usage.The parameter change from
batch=state.batchtograph_idx=state.graph_idxcorrectly demonstrates the updatedcalc_kTfunction signature in the tutorial.tests/models/conftest.py (2)
172-172: LGTM! Test updated for renamed attribute.The change from
sim_state.batchtosim_state.graph_idxcorrectly reflects the renamed attribute in theSimStateclass while maintaining the same test logic.
180-180: LGTM! Immutability assertion updated for renamed attribute.The assertion correctly verifies that the model doesn't mutate the
graph_idxattribute, maintaining the same test coverage with the updated naming.examples/scripts/3_Dynamics/3.12_MACE_NPT_Langevin.py (5)
71-71: LGTM! Parameter updated consistently in NVT loop.The change from
batch=state.batchtograph_idx=state.graph_idxin thecalc_kTfunction call correctly uses the updated API.
86-86: LGTM! Parameter updated consistently in NPT loop.The temperature calculation correctly uses the updated
graph_idxparameter naming.
95-95: LGTM! Kinetic energy calculation updated consistently.The
calc_kinetic_energyfunction call correctly uses the updatedgraph_idxparameter.
110-110: LGTM! Final temperature calculation updated.The final temperature calculation correctly uses the updated parameter naming.
120-120: LGTM! Final pressure calculation updated.The final kinetic energy calculation for pressure determination correctly uses the updated
graph_idxparameter.torch_sim/models/graphpes.py (1)
70-71: LGTM: Consistent terminology updateThe changes correctly update the iteration and masking logic to use the new graph-based terminology (
n_graphsandgraph_idx) while preserving the original functionality.examples/scripts/3_Dynamics/3.7_Lennard_Jones_NPT_Nose_Hoover.py (1)
126-126: LGTM: Function calls updated correctlyAll function calls to
calc_kTandcalc_kinetic_energyhave been properly updated to use the newgraph_idxparameter naming, consistent with the function signatures intorch_sim/quantities.py.Also applies to: 133-133, 148-148, 155-157
torch_sim/workflows/a2c.py (1)
733-736: LGTM: Tensor shapes updated correctlyThe logging tensor dimensions have been properly updated to use
state.n_graphsinstead ofstate.n_batches, maintaining the correct shape for tracking energy and stress over multiple systems during optimization.examples/scripts/7_Others/7.3_Batched_neighbor_list.py (1)
21-21: LGTM: Variable renaming and function calls updated correctlyThe variable has been correctly renamed from
batchtograph_idx, the comment accurately reflects the new tensor shape notation, and both neighbor list function calls have been properly updated to use the new variable name.Also applies to: 25-26, 29-29, 41-41
torch_sim/models/metatomic.py (2)
77-77: LGTM: Docstring updated correctlyThe docstring has been properly updated to reference "graph indices" instead of "batch indices", consistent with the new terminology.
203-203: LGTM: Mask logic updated correctlyThe system mask creation correctly uses
state.graph_idx == binstead ofstate.batch == b, maintaining the same logic while using the updated attribute name.tests/test_optimizers.py (7)
131-131: LGTM: Consistent parameter renamingThe parameter name change from
batchtograph_idxaligns with the systematic terminology update across the codebase.
232-232: LGTM: Comment update maintains clarityThe comment update from "per-batch dt" to "per-graph dt" maintains clarity while adopting the new terminology.
265-265: LGTM: Consistent indexing attribute renamingThe indexing operations correctly use
graph_idxinstead ofbatchwhile preserving the same selection logic.Also applies to: 268-268, 320-321
348-348: LGTM: Consistent constructor parameter updateThe SimState constructor parameter correctly uses
graph_idxinstead ofbatch.
432-432: LGTM: Property name updates are consistentThe property name changes from
n_batcheston_graphsalign with the updated API and maintain the same functionality.Also applies to: 456-456
459-459: LGTM: Consistent attribute usage in unique operationThe change from
state.batchtostate.graph_idxin thetorch.uniquecall maintains the same logic for counting unique graph indices.
877-878: LGTM: Error message terminology updatedThe error messages correctly use "graph" terminology instead of "batch", maintaining consistency with the new naming convention.
torch_sim/models/sevennet.py (2)
184-185: LGTM: Correct iteration and masking logic with new terminologyThe iteration range and batch masking logic correctly use
graph_idxinstead ofbatchwhile preserving the same functionality for processing each graph separately.
248-248: LGTM: Tensor sizing correctly updatedThe energy tensor sizing correctly uses
state.graph_idx.max().item() + 1to determine the number of graphs, maintaining the same logic with the updated attribute name.torch_sim/models/orb.py (3)
105-105: LGTM: Node counting logic correctly updatedThe
torch.bincountoperation correctly usesstate.graph_idxto count nodes per graph, maintaining the same functionality with the updated attribute name.
146-146: LGTM: System count calculation correctly updatedThe calculation of
n_systemscorrectly usesstate.graph_idx.max().item() + 1to determine the number of graphs, preserving the same logic.
160-160: LGTM: Batch masking logic correctly updatedThe batch mask creation correctly uses
state.graph_idx == ito select atoms belonging to each graph, maintaining the same selection logic.examples/tutorials/autobatching_tutorial.py (3)
252-252: LGTM: Property access correctly updated in tutorialThe change from
fire_state.n_batchestofire_state.n_graphscorrectly uses the updated property name while maintaining the same tracking functionality.
282-282: LGTM: Assertion correctly updated with new property nameThe assertion correctly uses
fire_state.n_graphs == 0to verify that all graphs have been processed, maintaining the same validation logic.
286-286: LGTM: Property demonstration correctly updatedThe property access correctly uses
fire_state.n_graphsinstead offire_state.n_batches, consistent with the API update.tests/test_io.py (6)
51-54: LGTM: Test assertions correctly updated for graph indexingThe assertions correctly validate
state.graph_idxshape and values instead ofstate.batch, maintaining the same test coverage with the updated attribute name.
68-69: LGTM: Single atoms test correctly updatedThe test correctly validates that
state.graph_idxhas the expected shape and that all atoms belong to graph 0 for a single structure.
83-87: LGTM: Multiple atoms test correctly updatedThe test correctly validates
state.graph_idxshape and values for multiple atom structures, ensuring proper graph index assignment.
175-179: LGTM: PhonopyAtoms test correctly updatedThe test correctly validates
state.graph_idxfor PhonopyAtoms conversion, maintaining the same validation logic with the updated attribute name.
240-244: LGTM: Round-trip test setup correctly updatedThe test correctly uses
torch.unique(sim_state.graph_idx)to count unique graphs and validates that the intermediate format preserves the correct number of structures.
253-253: LGTM: Round-trip validation correctly updatedThe assertion correctly validates that
graph_idxis preserved through round-trip conversion, maintaining the same test coverage with the updated attribute name.tests/test_autobatching.py (4)
118-121: LGTM! Consistent terminology update.The assertions correctly use the new
n_graphsandgraph_idxproperties instead of the old batch-related terminology. The logic remains unchanged.
475-482: LGTM! Graph-based convergence function is correctly updated.The convergence function properly uses
state.n_graphsandstate.graph_idxfor computing graph-wise maximum forces. The semantic meaning is preserved while using the clearer terminology.
517-538: LGTM! Optimal batch counting updated consistently.The variables tracking optimal number of graphs and the counting logic have been correctly updated to use graph-based terminology while maintaining the same computational logic.
564-577: LGTM! Convergence tensor creation uses correct graph count.The convergence tensor initialization correctly uses
state.n_graphsto create tensors with the appropriate size for tracking convergence per graph.torch_sim/neighbors.py (3)
769-769: LGTM! Parameter renamed for better semantic clarity.Renaming
batchtograph_idxmakes the function signature more descriptive, clearly indicating that this parameter represents the graph index for each atom.
787-787: LGTM! Docstring correctly updated.The parameter documentation properly reflects the new
graph_idxparameter name and maintains the same semantic meaning.
809-809: LGTM! Function usage updated consistently.The
torch.bincount(graph_idx)call correctly uses the renamed parameter while maintaining the same computational logic.torch_sim/autobatching.py (7)
237-237: LGTM! Print statement correctly updated.The diagnostic message now properly reports the number of graphs instead of batches, aligning with the new terminology.
296-297: LGTM! Variable naming updated consistently.The variable
n_graphsbetter represents what it's counting compared to the previous batch-related naming.
346-346: LGTM! Condition check uses correct property.The check for multiple graphs using
state.n_graphs > 1is semantically correct and consistent with the terminology update.
408-409: LGTM! Print statement terminology updated appropriately.The diagnostic message correctly refers to "graphs" instead of "batches" when describing the systems, making the output more accurate.
431-431: LGTM! Docstring updates improve clarity.The documentation correctly changes "per batch" to "per graph" in descriptions of the maximum memory scaler, making the parameter's purpose clearer.
Also applies to: 480-480, 719-719, 779-779
936-942: LGTM! Batch size determination variable naming updated.The variable
n_graphsis more descriptive than the previous naming and correctly represents what's being counted for memory estimation.
1022-1024: LGTM! Validation checks correctly updated.All validation checks now use
state.n_graphsconsistently, maintaining the same validation logic while using the clearer terminology.Also applies to: 1028-1030, 1060-1060
torch_sim/models/particle_life.py (2)
226-229: LGTM! Docstring updates align with graph terminology.The docstring changes correctly reflect the shift from batch-based to graph-based indexing for tensor shapes.
242-245: LGTM! Code changes correctly implement the batch→graph renaming.The attribute access and iteration logic remain identical - only the terminology has been updated from
batch/n_batchestograph_idx/n_graphs.tests/test_integrators.py (3)
112-112: LGTM! Function calls correctly updated to use graph_idx parameter.The
calc_kTfunction calls have been properly updated frombatch=state.batchtograph_idx=state.graph_idx, which aligns with the function signature shown in the relevant code snippets.Also applies to: 175-175, 216-216, 276-276
375-376: LGTM! Test function renamed consistently with terminology shift.The function name and docstring have been updated from "atoms_per_batch" to "atoms_per_graph" while preserving the same test logic.
392-392: LGTM! Constructor call updated with correct parameter name.The
NPTLangevinStateconstructor call has been correctly updated frombatch=tograph_idx=parameter.examples/scripts/7_Others/7.6_Compare_ASE_to_VV_FIRE.py (4)
144-144: LGTM! Property access correctly updated to use n_graphs.The property access has been systematically updated from
n_batcheston_graphsthroughout the optimization functions, maintaining the same logic flow.Also applies to: 171-171, 222-222, 233-233
332-332: LGTM! Variable names consistently renamed from batch to graph terminology.All variable names have been systematically updated:
all_batches_for_gd→all_graphs_for_gdbatch_idx→graph_idxconcatenated_batch_indices→concatenated_graph_indicesThe logic and data flow remain unchanged.
Also applies to: 337-337, 353-357, 397-397
364-364: LGTM! Print statements updated with consistent terminology.Error messages and print statements have been updated to use
graph_idxterminology while maintaining the same informational content.Also applies to: 378-378
416-416: LGTM! Constructor call updated with correct parameter name.The
GDStateconstructor call has been correctly updated frombatch=tograph_idx=parameter.tests/test_monte_carlo.py (3)
53-53: LGTM! Function calls correctly updated to use graph_idx parameter.The
validate_permutationfunction calls have been properly updated to usegraph_idxparameter, which aligns with the function signature shown in the relevant code snippets.Also applies to: 98-98, 108-108
67-69: LGTM! Test logic correctly updated with graph terminology.The comment and assertions have been updated to use
graph_idxwhile preserving the same validation logic for ensuring swaps occur within the same group.
150-163: LGTM! Monte Carlo validation updated with consistent terminology.The test code has been systematically updated to use
graph_idxfor grouping and validation while maintaining the same test logic for verifying atomic number distributions and graph assignments.tests/test_correlations.py (2)
34-37: LGTM! MockState attributes correctly updated to match SimState interface.The mock state attributes have been properly updated from
n_batches/batchton_graphs/graph_idx, aligning with the real SimState interface changes shown in the relevant code snippets.
40-42: LGTM! Comments updated with consistent terminology.The docstring and comments have been updated to use "graph" terminology while maintaining the same functional description.
torch_sim/models/morse.py (2)
359-363: LGTM! Tensor shape documentation correctly updated.The docstring updates properly reflect the new graph-based terminology, changing from
[n_batches]to[n_graphs]for energy and stress tensor shapes.
385-385: LGTM! Iteration logic correctly updated.The change from
state.n_batchestostate.n_graphsproperly reflects the new terminology while preserving the iteration logic.torch_sim/models/soft_sphere.py (2)
394-399: LGTM! Documentation properly updated for graph terminology.The tensor shapes and parameter descriptions correctly reflect the change from batch-based to graph-based indexing.
857-859: LGTM! Consistent error handling in multi-model.The SoftSphereMultiModel class correctly implements the same graph indexing logic and error handling as the base model.
torch_sim/models/fairchem.py (2)
352-353: LGTM! Graph index initialization correctly updated.The change from
state.batchtostate.graph_idxproperly reflects the new terminology, and the tensor initialization logic remains sound.
361-362: LGTM! Tensor operations correctly updated for graph indexing.The
torch.bincount(state.graph_idx)and subsequent tensor initialization properly use the new graph indexing scheme while preserving the original computational logic.tests/test_runners.py (4)
152-152: LGTM! Test trajectory file naming updated correctly.The change from
n_batcheston_graphsin the list comprehension properly reflects the new terminology while maintaining the test logic.
801-801: LGTM! Energy tensor shape assertion correctly updated.The assertion now properly checks against
n_graphsinstead ofn_batches, maintaining test correctness with the new terminology.
815-817: LGTM! Mock state graph indexing properly implemented.The mock state correctly uses
graph_idxwithtorch.repeat_interleaveandtorch.arange(n_graphs)to simulate the new graph indexing scheme.
933-935: LGTM! Controlled mock state maintains consistent graph indexing.The
ControlledMockStateclass properly implements the graph indexing pattern, ensuring test reliability with the new terminology.torch_sim/runners.py (1)
175-175: LGTM! Consistent batch-to-graph terminology updates.All changes correctly update progress bar operations and docstrings from
n_batches/(n_batches,)ton_graphs/(n_graphs,). The logic remains unchanged while improving terminology clarity.Also applies to: 197-197, 310-311, 346-347, 375-375, 437-437, 548-548, 571-571
torch_sim/integrators/npt.py (1)
34-42: LGTM! Comprehensive and consistent batch-to-graph renaming.This file demonstrates excellent systematic refactoring:
- All variable references updated from
batch/n_batchestograph_idx/n_graphs- Docstrings and type annotations properly updated with new tensor shapes
- Indexing operations consistently use
state.graph_idxinstead ofstate.batch- Algorithm logic remains unchanged while improving terminology clarity
The changes align perfectly with the PR objective of freeing up "batch" terminology for actual batching operations.
Also applies to: 89-96, 167-179, 231-236, 271-275, 318-321, 406-414, 477-481, 507-511, 547-550, 567-571, 646-647, 680-681, 730-736, 748-749, 852-856, 884-885, 887-893, 950-954, 962-964, 970-974, 990-996, 1013-1017, 1083-1085, 1148-1150, 1177-1178, 1188-1199, 1207-1221, 1269-1271, 1284-1296, 1409-1414, 1421-1425, 1434-1444, 1462-1463, 1477-1478, 1484-1485, 1541-1542, 1591-1610, 1620-1627
torch_sim/models/mace.py (1)
94-94: LGTM! Correct batch-to-graph interface updates with preserved model compatibility.The changes properly update the external interface to use
graph_idxterminology while correctly maintainingbatch=state.graph_idxin line 337 for the underlying MACE model interface. This approach:
- Updates the TorchSim interface to use clearer "graph" terminology
- Preserves compatibility with the existing MACE model API that expects "batch"
- Maintains consistency with the broader codebase renaming effort
The method rename from
setup_from_batchcould be considered for future cleanup, but the current approach correctly balances interface clarity with model compatibility.Also applies to: 115-115, 132-133, 189-195, 197-199, 208-209, 212-212, 282-287, 298-298, 308-308, 337-337
torch_sim/integrators/md.py (4)
24-24: Docstring updates look accurate.The terminology changes from "batch" to "graph" in the docstrings correctly reflect the new semantic understanding where each graph represents an independent system.
Also applies to: 27-27, 29-29, 34-34
54-54: Function signature and documentation updated correctly.The parameter rename from
batchtograph_idxand corresponding docstring updates are consistent with the codebase-wide terminology shift.Also applies to: 67-68, 82-83
150-152: Comment and function call updates are consistent.The comment update clarifies the graph-based approach, and the function call to
pbc_wrap_batchedcorrectly usesstate.graph_idx.
90-96: calculate_momenta renaming verified; no changes requiredThe mechanical rename from
batch→graphin torch_sim/integrators/md.py (lines 90–96) preserves the original center-of-mass removal logic. Existing tests in tests/test_integrators.py already cover both multi-atom and single-atom branches ofcalculate_momenta, so the flagged coverage gap is pre-existing and does not block approval.torch_sim/monte_carlo.py (4)
44-44: Function documentation and initial variable updates are correct.The docstring and initial parameter usage correctly reflect the shift from batch-based to graph-based terminology.
Also applies to: 54-54, 57-57
60-78: Complex graph-based swap generation logic preserved correctly.The renaming from batch-related variables to graph-related variables maintains the same computational logic:
graph_lengthsreplacesbatch_lengthsn_graphsreplacesn_batches- Graph indexing and range calculations are preserved
- The core algorithm for generating swaps within the same graph remains intact
The logic correctly ensures swaps only occur between atoms of different types within the same graph.
Also applies to: 82-103
127-141: Parameter and validation logic updated consistently.The function parameter change from
batchtograph_idxand the corresponding validation logic update correctly enforce that swaps must occur within the same graph.
236-236: State initialization and validation calls updated appropriately.The
graph_idxparameter usage in bothinit_swap_mc_stateand the validation call inswap_monte_carlo_stepcorrectly reflects the terminology change.Also applies to: 268-268
torch_sim/optimizers.py (6)
47-47: GDState docstring updates are accurate.The tensor shape annotations correctly reflect the graph-based indexing with
[n_graphs]replacing[n_batches].Also applies to: 50-50, 52-52
71-72: Gradient descent function updates preserve logic.The learning rate documentation and indexing logic correctly map from graph-based learning rates to per-atom rates using
state.graph_idx.Also applies to: 132-132, 134-134
273-273: Graph counting and indexing logic updated correctly.The changes to use
state.n_graphsandstate.graph_idxfor counting and indexing operations maintain the same computational semantics with the new terminology.Also applies to: 278-278, 321-321, 374-374
571-571: Dimension calculations updated consistently across optimizers.All optimizer initialization functions correctly use
state.n_graphsfor determining tensor dimensions, ensuring consistency across different optimization methods.Also applies to: 798-798, 1083-1083
1242-1242: FIRE step functions maintain correct graph-based power calculations.The
batched_vdotcalls correctly usestate.graph_idxfor computing per-graph power values, preserving the FIRE algorithm's velocity-force dot product calculations.Also applies to: 1369-1369, 1460-1460
1280-1280: Graph iteration loops updated correctly.The for loops that iterate over graphs correctly use
range(n_graphs)and maintain the same computational logic for per-graph operations in both Velocity Verlet and ASE FIRE implementations.Also applies to: 1344-1344, 1650-1650
add .vscode to gitignore rename per_batch to per_graph wip convert many batches to graphs convert more batches to graphs rename all of batch indices rename all of batch indices wip fix tests fix more tests more test renaming more test renaming again fix test_deform_grad_batched rename more batch to use graphs update mock state to use graphs precommit fix bug rename in calculate_momenta fix more precommit rename more batch to graphs more renames more renaming fix deprecated text in docstring added test for deprecated batch properties in simstate rename batch to graph more renames cleanup minor typo
curtischong
force-pushed
the
rename-mdstate
branch
from
c555b9d to
0aa7991
Compare
|
i'm okay with this but I would probably prefer the word Also need to update the version, whilst this is breaking API change I don't think it really looks like a MAJOR version increment so would bump a MINOR version. |
|
closing since #217 supersedes this |
Summary
This change is quite involved. But I didn't want to split it up into multiple parts because "batch" is so pervasive within the codebase. But it feels like the right thing to do. Notable, I changed these interfaces in the SimState:
The changes to getitem and init will cause breaking changes, but I feel like it is worth it while torchsim's adoption is currently minimal.
Notes of my changes
Checklist
Before a pull request can be merged, the following items must be checked:
Summary by CodeRabbit
Refactor
batch→graph_idx,n_batches→n_graphs) without changing logic or control flow.Tests
Style
Chores
.gitignoreto exclude the.vscode/directory.