Skip to content

⚡️ Speed up function create_array by 11% #780

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
codeflash-ai wants to merge 1 commit into
base: main
Choose a base branch
from
Open

⚡️ Speed up function create_array by 11% #780

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
17 changes: 14 additions & 3 deletions inference/core/workflows/execution_engine/v1/executor/output_constructor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -316,10 +316,21 @@ def create_array(indices: np.ndarray) -> Optional[list]:
if indices.size == 0:
return None
result = []
max_idx = indices[:, 0].max() + 1
# Use np.unique to avoid searching 'max' if indices[:,0] is not zero-based and for direct iteration
first_col = indices[:, 0]
unique_idxs = np.unique(first_col)
idx_map = {idx: [] for idx in unique_idxs}
# Efficiently group rows by their first column index without repeated slicing
for i, idx in enumerate(first_col):
idx_map[idx].append(i)
max_idx = unique_idxs.max() + 1
for idx in range(max_idx):
idx_selector = indices[:, 0] == idx
indices_subset = indices[idx_selector][:, 1:]
if idx in idx_map:
idx_indices = idx_map[idx]
# Use array slicing instead of boolean indexing
indices_subset = indices[idx_indices][:, 1:]
else:
indices_subset = np.empty((0, indices.shape[1] - 1), dtype=indices.dtype)
inner_array = create_array(indices_subset)
if (
inner_array is None
Expand Down
Loading