Skip to content

Backport fixes from PR #248 #249

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
lohedges merged 1 commit into from
Oct 15, 2026
Merged

Backport fixes from PR #248 #249

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 14 additions & 5 deletions src/somd2/runner/_base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2373,6 +2373,7 @@ def _checkpoint(
is_final_block=False,
context=None,
gcmc_sampler=None,
num_energy_rows=None,
):
"""
Save a checkpoint file.
Expand Down Expand Up @@ -2401,6 +2402,10 @@ def _checkpoint(
is_final_block: bool
Whether this is the final block of the simulation.

num_energy_rows: int
The number of energy samples collected since the last checkpoint.
If None, then a full block is assumed.

Returns
-------

Expand All @@ -2427,6 +2432,13 @@ def _checkpoint(
# parquet-related work in this case.
is_post_equilibration = block == -1

def new_rows(df):
# The rows sampled since the last checkpoint. Sliced from the
# start, since a count of zero must select no rows.
if num_energy_rows is None:
return df.iloc[-self._energy_per_block :]
return df.iloc[len(df) - num_energy_rows :]

# Get the energy trajectory.
if not is_post_equilibration:
df = system.energy_trajectory(to_alchemlyb=True, energy_unit="kT")
Expand Down Expand Up @@ -2518,7 +2530,7 @@ def _checkpoint(
# yet (e.g. checkpoint_frequency=0), create one from scratch.
_energy_traj = self._filenames[index]["energy_traj"]
if _Path(_energy_traj).exists():
_parquet_append(_energy_traj, df.iloc[-self._energy_per_block :])
_parquet_append(_energy_traj, new_rows(df))
else:
_dataframe_to_parquet(df, metadata=metadata, filename=_energy_traj)

Expand Down Expand Up @@ -2558,10 +2570,7 @@ def _checkpoint(
):
_dataframe_to_parquet(df, metadata=metadata, filename=filename)
else:
_parquet_append(
filename,
df.iloc[-self._energy_per_block :],
)
_parquet_append(filename, new_rows(df))

if not is_post_equilibration:
self._flush_energy_components(index)
Expand Down
87 changes: 81 additions & 6 deletions src/somd2/runner/_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -79,6 +79,61 @@ def __init__(self, system, config):
_logger.error(msg)
raise ValueError(msg)

from math import isclose

def is_multiple(value, interval):
ratio = (value / interval).value()
return isclose(ratio, round(ratio), abs_tol=1e-4)

# Dynamics are run in whole integration steps.
for option in (
"runtime",
"checkpoint_frequency",
"energy_frequency",
"frame_frequency",
"gcmc_frequency",
"terminal_flip_frequency",
):
value = getattr(config, option)
if value is not None and not is_multiple(value, config.timestep):
msg = f"'{option}' must be a multiple of 'timestep'."
_logger.error(msg)
raise ValueError(msg)

# Blocks are split into sub-blocks between Monte Carlo moves, or energy
# component samples, so they must hold a whole number of them.
is_checkpointing = config.checkpoint_frequency.value() > 0.0
intervals = {}
if config.gcmc:
intervals["gcmc_frequency"] = (
config.gcmc_frequency or config.energy_frequency
)
if config.terminal_flip_frequency is not None:
intervals["terminal_flip_frequency"] = config.terminal_flip_frequency
if config.save_energy_components:
intervals["energy_frequency"] = config.energy_frequency
options = (
("runtime", "checkpoint_frequency") if is_checkpointing else ("runtime",)
)
for name, interval in intervals.items():
for option in options:
if not is_multiple(getattr(config, option), interval):
msg = f"'{option}' must be a multiple of '{name}'."
_logger.error(msg)
raise ValueError(msg)

if is_checkpointing:
# A final partial block doesn't perform GCMC moves.
if config.gcmc and not is_multiple(
config.runtime, config.checkpoint_frequency
):
msg = (
"'runtime' must be a multiple of 'checkpoint_frequency' "
"for GCMC simulations."
)
_logger.error(msg)
raise ValueError(msg)

# Call the base class constructor.
super().__init__(system, config)

Expand Down Expand Up @@ -906,6 +961,10 @@ def generate_lam_vals(lambda_base, increment=0.001):
# Store the start time.
start = _timer()

# The number of energy samples recorded when the last checkpoint was
# written, so that each checkpoint appends only the new ones.
num_samples = len(dynamics._d._energy_trajectory)

# Run the simulation, checkpointing in blocks.
if checkpoint_frequency.value() > 0.0:
# Calculate the number of blocks and the remainder time.
Expand Down Expand Up @@ -989,7 +1048,12 @@ def generate_lam_vals(lambda_base, increment=0.001):
else None
)

while runtime < checkpoint_frequency:
# Count sub-blocks rather than compare times, since a
# rounding error can otherwise add one.
num_subblocks = round(
(checkpoint_frequency / block_size).value()
)
while flip_counter < num_subblocks:
needs_pre_run_snapshot = False

# GCMC move.
Expand Down Expand Up @@ -1071,9 +1135,7 @@ def generate_lam_vals(lambda_base, increment=0.001):
self._save_energy_components(
index,
dynamics.context(),
(block * checkpoint_frequency + ec_elapsed).to(
"ns"
),
dynamics.current_time().to("ns"),
)

else:
Expand Down Expand Up @@ -1136,10 +1198,13 @@ def generate_lam_vals(lambda_base, increment=0.001):
is_final_block=is_final_block,
context=dynamics.context(),
gcmc_sampler=gcmc_sampler,
num_energy_rows=len(dynamics._d._energy_trajectory)
- num_samples,
)

if error is not None:
raise error
num_samples = len(dynamics._d._energy_trajectory)

# Save sampler statistics alongside the checkpoint.
self._save_sampler_stats(
Expand Down Expand Up @@ -1214,8 +1279,12 @@ def generate_lam_vals(lambda_base, increment=0.001):
):
dynamics.randomise_velocities()

# Run the remaining time as a whole number of steps, since
# Sire truncates a time a rounding error short of a step.
dynamics.run(
rem,
round(
(rem * checkpoint_frequency / self._config.timestep).value()
),
energy_frequency=self._config.energy_frequency,
frame_frequency=self._config.frame_frequency,
lambda_windows=lambda_array,
Expand Down Expand Up @@ -1268,6 +1337,8 @@ def generate_lam_vals(lambda_base, increment=0.001):
is_final_block=True,
context=dynamics.context(),
gcmc_sampler=gcmc_sampler,
num_energy_rows=len(dynamics._d._energy_trajectory)
- num_samples,
)

if error is not None:
Expand Down Expand Up @@ -1346,7 +1417,10 @@ def generate_lam_vals(lambda_base, increment=0.001):
)
time_base = self._config.runtime - time

while runtime < time:
# Count sub-blocks rather than compare times, since a
# rounding error can otherwise add one.
num_subblocks = round((time / block_size).value())
while flip_counter < num_subblocks:
needs_pre_run_snapshot = False

# GCMC move.
Expand Down Expand Up @@ -1481,6 +1555,7 @@ def generate_lam_vals(lambda_base, increment=0.001):
is_final_block=True,
context=dynamics.context(),
gcmc_sampler=gcmc_sampler,
num_energy_rows=len(dynamics._d._energy_trajectory) - num_samples,
)

if error is not None:
Expand Down
4 changes: 2 additions & 2 deletions tests/runner/test_terminal_flip.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -397,8 +397,8 @@ def test_runner_validation_frequency_multiple(ethane_methanol):
output_directory=tmpdir,
platform="cpu",
num_lambda=2,
energy_frequency="4fs",
terminal_flip_frequency="3fs", # not a multiple of 4fs
energy_frequency="8fs",
terminal_flip_frequency="20fs", # not a multiple of 8fs
)
with pytest.raises(
ValueError, match="must be a multiple of 'energy_frequency'"
Expand Down