Skip to content

Include initial snapshot in exports #3

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

Open
lechevaa wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Include initial snapshot in exports #3

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
14 changes: 9 additions & 5 deletions notebooks/spe11b_task_1.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@
"- `rhoG [kg/m3]`\n",
"- `rhoL [kg/m3]`\n",
"- `tmCO2 [kg]`\n",
"- `temp [°C]`\n",
"- `temp [\u00b0C]`\n",
"\n",
"The surrogate predicts the state variables forward in time and keeps the coordinates fixed from the template CSV.\n",
"\n",
Expand Down Expand Up @@ -135,7 +135,7 @@
"\n",
"# Train on one participant from the first 50 years only.\n",
"train_participant = participants[0]\n",
"export_participant = participants[3]\n",
"export_participant = participants[4]\n",
"train_dataset = SpatialMapTransitionDataset(\n",
" data_root,\n",
" participants=[train_participant],\n",
Expand Down Expand Up @@ -454,7 +454,7 @@
},
{
"cell_type": "code",
"execution_count": 9,
"execution_count": null,
"id": "329316dc",
"metadata": {},
"outputs": [],
Expand All @@ -479,9 +479,13 @@
"ar_summaries = []\n",
"ar_feature_names = [name for name in field_names if name not in ['x [m]', 'z [m]', 'pressure [Pa]']]\n",
"ar_feature_names = [name for name in ar_feature_names if name in physical_fields]\n",
"ar_feature_traces = {name: [] for name in ar_feature_names}\n",
"ar_feature_traces = {name: [0.0] for name in ar_feature_names}\n",
"\n",
"current_snapshot = export_series.snapshots[0]\n",
"initial_output_path = ar_root / export_series.participant / current_snapshot.path.name\n",
"save_spatial_map_csv_data(current_snapshot, initial_output_path)\n",
"ar_outputs.append(initial_output_path)\n",
"ar_summaries.append({'target_time_years': current_snapshot.time_years, 'pressure [Pa]': 0.0})\n",
"for target_snapshot in export_series.snapshots[1:]:\n",
" predicted_data = predict_autoregressive_data(current_snapshot, target_snapshot)\n",
" output_path = ar_root / export_series.participant / target_snapshot.path.name\n",
Expand Down Expand Up @@ -625,4 +629,4 @@
},
"nbformat": 4,
"nbformat_minor": 5
}
}
27 changes: 20 additions & 7 deletions src/ml4gcs/data/export.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,19 +50,22 @@ def save_spatial_map_csv(
def save_spatial_map_csv_data(
template: SpatialMapSnapshot,
output_path: Path | str,
data: np.ndarray,
data: np.ndarray | None = None,
) -> Path:
"""Write a spatial-map CSV using a full row matrix."""

output_path = Path(output_path)
output_path.parent.mkdir(parents=True, exist_ok=True)

data = np.asarray(data, dtype=template.data.dtype)
if data.shape != template.data.shape:
raise ValueError(
"Predicted data shape does not match the template shape: "
f"{data.shape} vs {template.data.shape}"
)
if data is None:
data = template.data.copy()
else:
data = np.asarray(data, dtype=template.data.dtype)
if data.shape != template.data.shape:
raise ValueError(
"Predicted data shape does not match the template shape: "
f"{data.shape} vs {template.data.shape}"
)

with output_path.open("w", encoding="utf-8", newline="") as handle:
handle.write("# " + ", ".join(template.columns) + "\n")
Expand Down Expand Up @@ -109,6 +112,12 @@ def export_next_step_timeline(
output_paths: list[Path] = []
snapshots = participant_series.snapshots

if snapshots:
first_snapshot = snapshots[0]
output_paths.append(
save_spatial_map_csv_data(first_snapshot, output_root / first_snapshot.path.name)
)

for input_snapshot, target_snapshot in zip(snapshots[:-1], snapshots[1:], strict=True):
pressure_grid = predict_pressure(input_snapshot, target_snapshot)
output_path = output_root / target_snapshot.path.name
Expand All @@ -128,6 +137,10 @@ def export_next_step_timeline_data(
output_paths: list[Path] = []
snapshots = participant_series.snapshots

if snapshots:
first_snapshot = snapshots[0]
output_paths.append(save_spatial_map_csv_data(first_snapshot, output_root / first_snapshot.path.name))

for input_snapshot, target_snapshot in zip(snapshots[:-1], snapshots[1:], strict=True):
predicted = predict_data(input_snapshot, target_snapshot)
output_path = output_root / target_snapshot.path.name
Expand Down