Skip to content

Fix: COOMatrix float data copy size in GPU path #758

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

Fix: COOMatrix float data copy size in GPU path #758

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
2 changes: 1 addition & 1 deletion implicit/gpu/matrix.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -270,7 +270,7 @@ COOMatrix::COOMatrix(int rows, int cols, int nonzeros, const int *row_,

CHECK_CUDA(cudaMallocManaged(&data, nonzeros * sizeof(float)));
CHECK_CUDA(
cudaMemcpy(data, data_, nonzeros * sizeof(int), cudaMemcpyHostToDevice));
cudaMemcpy(data, data_, nonzeros * sizeof(float), cudaMemcpyHostToDevice));
}

COOMatrix::~COOMatrix() {
Expand Down
12 changes: 12 additions & 0 deletions tests/gpu_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
import pytest
from numpy.testing import assert_allclose, assert_array_equal
from recommender_base_test import get_checker_board
from scipy.sparse import coo_matrix

import implicit

Expand Down Expand Up @@ -81,3 +82,14 @@ def test_cpu_gpu_conversion(model_class, from_gpu):
rtol=1e-3,
atol=1e-3,
)


@pytest.mark.skipif(not implicit.gpu.HAS_CUDA, reason="needs cuda build")
def test_coo_matrix_copies_float_data():
row = np.array([0, 1, 2, 0, 1, 2], dtype=np.int32)
col = np.array([0, 1, 2, 1, 2, 0], dtype=np.int32)
data = np.array([1.5, 2.7, 3.14159, 0.1, 1000.5, -1.5], dtype=np.float32)

matrix = implicit.gpu.COOMatrix(coo_matrix((data, (row, col)), shape=(3, 3)))

assert matrix is not None