Skip to content

Commit 64c42c3

Browse files
authored
fix excessive memory usage
1 parent 53a4a7f commit 64c42c3

File tree

1 file changed

+5
-1
lines changed

1 file changed

+5
-1
lines changed

src/database_interactions.py

Lines changed: 5 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -408,13 +408,17 @@ def create_database(self, texts, embeddings):
408408
with open(self.ROOT_DIRECTORY / "config.yaml", 'r', encoding='utf-8') as config_file:
409409
config_data = yaml.safe_load(config_file)
410410

411-
# precompute vectors, then write DB
411+
# precompute vectors
412412
vectors = embeddings.embed_documents(all_texts)
413413
text_embed_pairs = [
414414
(txt, np.asarray(vec, dtype=np.float32))
415415
for txt, vec in zip(all_texts, vectors)
416416
]
417417

418+
# IMMEDIATE CLEANUP - free ~50-75% of memory
419+
del all_texts, vectors
420+
gc.collect()
421+
418422
TileDB.from_embeddings(
419423
text_embeddings=text_embed_pairs,
420424
embedding=embeddings,

0 commit comments

Comments
 (0)