Skip to content

[Repo Assist] perf(gcm): avoid pre-allocated DataFrame in compute_data_from_noise and compute_noise_from_data #1561

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

Draft
github-actions wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

[Repo Assist] perf(gcm): avoid pre-allocated DataFrame in compute_data_from_noise and compute_noise_from_data #1561

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: 8 additions & 6 deletions dowhy/gcm/_noise.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
from typing import Any, Callable, List, Optional, Tuple
from typing import Any, Callable, Dict, List, Optional, Tuple

import networkx as nx
import numpy as np
Expand All @@ -19,28 +19,30 @@ def compute_data_from_noise(causal_model: StructuralCausalModel, noise_data: pd.
validate_causal_dag(causal_model.graph)

sorted_nodes = list(nx.topological_sort(causal_model.graph))
data = pd.DataFrame(np.empty((noise_data.shape[0], len(sorted_nodes))), columns=sorted_nodes)
data: Dict[Any, np.ndarray] = {}

for node in sorted_nodes:
if is_root_node(causal_model.graph, node):
data[node] = noise_data[node].to_numpy().squeeze()
else:
parents = get_ordered_predecessors(causal_model.graph, node)
data[node] = (
causal_model.causal_mechanism(node)
.evaluate(
data[get_ordered_predecessors(causal_model.graph, node)].to_numpy(), noise_data[node].to_numpy()
np.column_stack([data[p] for p in parents]),
noise_data[node].to_numpy(),
)
.squeeze()
)

return data
return pd.DataFrame(data, columns=sorted_nodes)


def compute_noise_from_data(causal_model: InvertibleStructuralCausalModel, observed_data: pd.DataFrame) -> pd.DataFrame:
validate_causal_dag(causal_model.graph)

sorted_noise = list(nx.topological_sort(causal_model.graph))
noise = pd.DataFrame(np.empty((observed_data.shape[0], len(sorted_noise))), columns=sorted_noise)
noise: Dict[Any, np.ndarray] = {}

for node in sorted_noise:
if is_root_node(causal_model.graph, node):
Expand All @@ -52,7 +54,7 @@ def compute_noise_from_data(causal_model: InvertibleStructuralCausalModel, obser
.squeeze()
)

return noise
return pd.DataFrame(noise, columns=sorted_noise)


def get_noise_dependent_function(
Expand Down