reduce bm25 memory usage

raphaelsty · Jun 2, 2024 · b36f8ae · b36f8ae
1 parent e99e0ac
commit b36f8ae
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/pyproject.toml b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "lenlp"
-version = "1.1.0"
+version = "1.1.1"
 description = "Natural Language Processing library for Python from Rust."
 
 authors = [

diff --git a/python/lenlp/sparse/bm25_vectorizer.py b/python/lenlp/sparse/bm25_vectorizer.py
@@ -109,7 +109,7 @@ def _transform(self, matrix: csr_matrix) -> csr_matrix:
             )
         )
 
-        denominator = matrix.copy().tocsc()
+        denominator = matrix.tocsc()
         denominator.data += np.take(a=regularization, indices=denominator.indices)
         matrix.data = (
             (matrix.data * (self.k1 + 1)) / denominator.tocsr().data