Skip to content

ARROW-6671: [C++][Python] Use more consistent names for sparse tensor items #5605

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

Closed
wants to merge 1 commit into from
Closed

ARROW-6671: [C++][Python] Use more consistent names for sparse tensor items #5605

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
18 changes: 9 additions & 9 deletions cpp/src/arrow/ipc/read_write_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1151,8 +1151,8 @@ class TestSparseTensorRoundTrip : public ::testing::Test, public IpcTestFixture
void SetUp() { IpcTestFixture::SetUp(); }
void TearDown() { IpcTestFixture::TearDown(); }

void CheckSparseTensorRoundTrip(const SparseTensorCOO& sparse_tensor);
void CheckSparseTensorRoundTrip(const SparseTensorCSR& sparse_tensor);
void CheckSparseTensorRoundTrip(const SparseCOOTensor& sparse_tensor);
void CheckSparseTensorRoundTrip(const SparseCSRMatrix& sparse_tensor);

protected:
std::shared_ptr<SparseCOOIndex> MakeSparseCOOIndex(
Expand All @@ -1166,19 +1166,19 @@ class TestSparseTensorRoundTrip : public ::testing::Test, public IpcTestFixture
}

template <typename ValueType>
std::shared_ptr<SparseTensorCOO> MakeSparseTensorCOO(
std::shared_ptr<SparseCOOTensor> MakeSparseCOOTensor(
const std::shared_ptr<SparseCOOIndex>& si, std::vector<ValueType>& sparse_values,
const std::vector<int64_t>& shape,
const std::vector<std::string>& dim_names = {}) const {
auto data = Buffer::Wrap(sparse_values);
return std::make_shared<SparseTensorCOO>(si, CTypeTraits<ValueType>::type_singleton(),
return std::make_shared<SparseCOOTensor>(si, CTypeTraits<ValueType>::type_singleton(),
data, shape, dim_names);
}
};

template <typename IndexValueType>
void TestSparseTensorRoundTrip<IndexValueType>::CheckSparseTensorRoundTrip(
const SparseTensorCOO& sparse_tensor) {
const SparseCOOTensor& sparse_tensor) {
const auto& type = checked_cast<const FixedWidthType&>(*sparse_tensor.type());
const int elem_size = type.bit_width() / 8;
const int index_elem_size = sizeof(typename IndexValueType::c_type);
Expand Down Expand Up @@ -1214,7 +1214,7 @@ void TestSparseTensorRoundTrip<IndexValueType>::CheckSparseTensorRoundTrip(

template <typename IndexValueType>
void TestSparseTensorRoundTrip<IndexValueType>::CheckSparseTensorRoundTrip(
const SparseTensorCSR& sparse_tensor) {
const SparseCSRMatrix& sparse_tensor) {
const auto& type = checked_cast<const FixedWidthType&>(*sparse_tensor.type());
const int elem_size = type.bit_width() / 8;
const int index_elem_size = sizeof(typename IndexValueType::c_type);
Expand Down Expand Up @@ -1291,7 +1291,7 @@ TYPED_TEST_P(TestSparseTensorRoundTrip, WithSparseCOOIndexRowMajor) {
std::vector<int64_t> shape = {2, 3, 4};
std::vector<std::string> dim_names = {"foo", "bar", "baz"};
std::vector<int64_t> values = {1, 2, 3, 4, 5, 6, 11, 12, 13, 14, 15, 16};
auto st = this->MakeSparseTensorCOO(si, values, shape, dim_names);
auto st = this->MakeSparseCOOTensor(si, values, shape, dim_names);

this->CheckSparseTensorRoundTrip(*st);
}
Expand Down Expand Up @@ -1334,7 +1334,7 @@ TYPED_TEST_P(TestSparseTensorRoundTrip, WithSparseCOOIndexColumnMajor) {
std::vector<int64_t> shape = {2, 3, 4};
std::vector<std::string> dim_names = {"foo", "bar", "baz"};
std::vector<int64_t> values = {1, 2, 3, 4, 5, 6, 11, 12, 13, 14, 15, 16};
auto st = this->MakeSparseTensorCOO(si, values, shape, dim_names);
auto st = this->MakeSparseCOOTensor(si, values, shape, dim_names);

this->CheckSparseTensorRoundTrip(*st);
}
Expand All @@ -1353,7 +1353,7 @@ TYPED_TEST_P(TestSparseTensorRoundTrip, WithSparseCSRIndex) {

auto data = Buffer::Wrap(values);
NumericTensor<Int64Type> t(data, shape, {}, dim_names);
SparseTensorImpl<SparseCSRIndex> st(t, TypeTraits<IndexValueType>::type_singleton());
SparseCSRMatrix st(t, TypeTraits<IndexValueType>::type_singleton());

this->CheckSparseTensorRoundTrip(st);
}
Expand Down
6 changes: 2 additions & 4 deletions cpp/src/arrow/ipc/reader.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -920,8 +920,7 @@ Status MakeSparseTensorWithSparseCOOIndex(
const std::vector<std::string>& dim_names,
const std::shared_ptr<SparseCOOIndex>& sparse_index, int64_t non_zero_length,
const std::shared_ptr<Buffer>& data, std::shared_ptr<SparseTensor>* out) {
*out = std::make_shared<SparseTensorImpl<SparseCOOIndex>>(sparse_index, type, data,
shape, dim_names);
*out = std::make_shared<SparseCOOTensor>(sparse_index, type, data, shape, dim_names);
return Status::OK();
}

Expand All @@ -930,8 +929,7 @@ Status MakeSparseTensorWithSparseCSRIndex(
const std::vector<std::string>& dim_names,
const std::shared_ptr<SparseCSRIndex>& sparse_index, int64_t non_zero_length,
const std::shared_ptr<Buffer>& data, std::shared_ptr<SparseTensor>* out) {
*out = std::make_shared<SparseTensorImpl<SparseCSRIndex>>(sparse_index, type, data,
shape, dim_names);
*out = std::make_shared<SparseCSRMatrix>(sparse_index, type, data, shape, dim_names);
return Status::OK();
}

Expand Down
24 changes: 12 additions & 12 deletions cpp/src/arrow/python/numpy_convert.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -315,7 +315,7 @@ static Status SparseTensorDataToNdarray(const SparseTensor& sparse_tensor,
return Status::OK();
}

Status SparseTensorCOOToNdarray(const std::shared_ptr<SparseTensorCOO>& sparse_tensor,
Status SparseCOOTensorToNdarray(const std::shared_ptr<SparseCOOTensor>& sparse_tensor,
PyObject* base, PyObject** out_data,
PyObject** out_coords) {
const auto& sparse_index = arrow::internal::checked_cast<const SparseCOOIndex&>(
Expand All @@ -335,7 +335,7 @@ Status SparseTensorCOOToNdarray(const std::shared_ptr<SparseTensorCOO>& sparse_t
return Status::OK();
}

Status SparseTensorCSRToNdarray(const std::shared_ptr<SparseTensorCSR>& sparse_tensor,
Status SparseCSRMatrixToNdarray(const std::shared_ptr<SparseCSRMatrix>& sparse_tensor,
PyObject* base, PyObject** out_data,
PyObject** out_indptr, PyObject** out_indices) {
const auto& sparse_index = arrow::internal::checked_cast<const SparseCSRIndex&>(
Expand All @@ -358,10 +358,10 @@ Status SparseTensorCSRToNdarray(const std::shared_ptr<SparseTensorCSR>& sparse_t
return Status::OK();
}

Status NdarraysToSparseTensorCOO(MemoryPool* pool, PyObject* data_ao, PyObject* coords_ao,
Status NdarraysToSparseCOOTensor(MemoryPool* pool, PyObject* data_ao, PyObject* coords_ao,
const std::vector<int64_t>& shape,
const std::vector<std::string>& dim_names,
std::shared_ptr<SparseTensorCOO>* out) {
std::shared_ptr<SparseCOOTensor>* out) {
if (!PyArray_Check(data_ao) || !PyArray_Check(coords_ao)) {
return Status::TypeError("Did not pass ndarray object");
}
Expand All @@ -383,10 +383,10 @@ Status NdarraysToSparseTensorCOO(MemoryPool* pool, PyObject* data_ao, PyObject*
return Status::OK();
}

Status NdarraysToSparseTensorCSR(MemoryPool* pool, PyObject* data_ao, PyObject* indptr_ao,
Status NdarraysToSparseCSRMatrix(MemoryPool* pool, PyObject* data_ao, PyObject* indptr_ao,
PyObject* indices_ao, const std::vector<int64_t>& shape,
const std::vector<std::string>& dim_names,
std::shared_ptr<SparseTensorCSR>* out) {
std::shared_ptr<SparseCSRMatrix>* out) {
if (!PyArray_Check(data_ao) || !PyArray_Check(indptr_ao) ||
!PyArray_Check(indices_ao)) {
return Status::TypeError("Did not pass ndarray object");
Expand All @@ -412,15 +412,15 @@ Status NdarraysToSparseTensorCSR(MemoryPool* pool, PyObject* data_ao, PyObject*
return Status::OK();
}

Status TensorToSparseTensorCOO(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseTensorCOO>* out) {
*out = std::make_shared<SparseTensorCOO>(*tensor);
Status TensorToSparseCOOTensor(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseCOOTensor>* out) {
*out = std::make_shared<SparseCOOTensor>(*tensor);
return Status::OK();
}

Status TensorToSparseTensorCSR(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseTensorCSR>* out) {
*out = std::make_shared<SparseTensorCSR>(*tensor);
Status TensorToSparseCSRMatrix(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseCSRMatrix>* out) {
*out = std::make_shared<SparseCSRMatrix>(*tensor);
return Status::OK();
}

Expand Down
22 changes: 11 additions & 11 deletions cpp/src/arrow/python/numpy_convert.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,30 +62,30 @@ ARROW_PYTHON_EXPORT Status TensorToNdarray(const std::shared_ptr<Tensor>& tensor
PyObject* base, PyObject** out);

ARROW_PYTHON_EXPORT Status
SparseTensorCOOToNdarray(const std::shared_ptr<SparseTensorCOO>& sparse_tensor,
SparseCOOTensorToNdarray(const std::shared_ptr<SparseCOOTensor>& sparse_tensor,
PyObject* base, PyObject** out_data, PyObject** out_coords);

ARROW_PYTHON_EXPORT Status SparseTensorCSRToNdarray(
const std::shared_ptr<SparseTensorCSR>& sparse_tensor, PyObject* base,
ARROW_PYTHON_EXPORT Status SparseCSRMatrixToNdarray(
const std::shared_ptr<SparseCSRMatrix>& sparse_tensor, PyObject* base,
PyObject** out_data, PyObject** out_indptr, PyObject** out_indices);

ARROW_PYTHON_EXPORT Status NdarraysToSparseTensorCOO(
ARROW_PYTHON_EXPORT Status NdarraysToSparseCOOTensor(
MemoryPool* pool, PyObject* data_ao, PyObject* coords_ao,
const std::vector<int64_t>& shape, const std::vector<std::string>& dim_names,
std::shared_ptr<SparseTensorCOO>* out);
std::shared_ptr<SparseCOOTensor>* out);

ARROW_PYTHON_EXPORT Status NdarraysToSparseTensorCSR(
ARROW_PYTHON_EXPORT Status NdarraysToSparseCSRMatrix(
MemoryPool* pool, PyObject* data_ao, PyObject* indptr_ao, PyObject* indices_ao,
const std::vector<int64_t>& shape, const std::vector<std::string>& dim_names,
std::shared_ptr<SparseTensorCSR>* out);
std::shared_ptr<SparseCSRMatrix>* out);

ARROW_PYTHON_EXPORT Status
TensorToSparseTensorCOO(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseTensorCOO>* csparse_tensor);
TensorToSparseCOOTensor(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseCOOTensor>* csparse_tensor);

ARROW_PYTHON_EXPORT Status
TensorToSparseTensorCSR(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseTensorCSR>* csparse_tensor);
TensorToSparseCSRMatrix(const std::shared_ptr<Tensor>& tensor,
std::shared_ptr<SparseCSRMatrix>* csparse_tensor);

} // namespace py
} // namespace arrow
Expand Down
32 changes: 16 additions & 16 deletions cpp/src/arrow/python/pyarrow.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,42 +132,42 @@ PyObject* wrap_tensor(const std::shared_ptr<Tensor>& tensor) {
return ::pyarrow_wrap_tensor(tensor);
}

bool is_sparse_tensor_csr(PyObject* sparse_tensor) {
return ::pyarrow_is_sparse_tensor_csr(sparse_tensor) != 0;
bool is_sparse_csr_matrix(PyObject* sparse_tensor) {
return ::pyarrow_is_sparse_csr_matrix(sparse_tensor) != 0;
}

Status unwrap_sparse_tensor_csr(PyObject* sparse_tensor,
std::shared_ptr<SparseTensorCSR>* out) {
*out = ::pyarrow_unwrap_sparse_tensor_csr(sparse_tensor);
Status unwrap_sparse_csr_matrix(PyObject* sparse_tensor,
std::shared_ptr<SparseCSRMatrix>* out) {
*out = ::pyarrow_unwrap_sparse_csr_matrix(sparse_tensor);
if (*out) {
return Status::OK();
} else {
return Status::Invalid(
"Could not unwrap SparseTensorCSR from the passed Python object.");
"Could not unwrap SparseCSRMatrix from the passed Python object.");
}
}

PyObject* wrap_sparse_tensor_csr(const std::shared_ptr<SparseTensorCSR>& sparse_tensor) {
return ::pyarrow_wrap_sparse_tensor_csr(sparse_tensor);
PyObject* wrap_sparse_csr_matrix(const std::shared_ptr<SparseCSRMatrix>& sparse_tensor) {
return ::pyarrow_wrap_sparse_csr_matrix(sparse_tensor);
}

bool is_sparse_tensor_coo(PyObject* sparse_tensor) {
return ::pyarrow_is_sparse_tensor_coo(sparse_tensor) != 0;
bool is_sparse_coo_tensor(PyObject* sparse_tensor) {
return ::pyarrow_is_sparse_coo_tensor(sparse_tensor) != 0;
}

Status unwrap_sparse_tensor_coo(PyObject* sparse_tensor,
std::shared_ptr<SparseTensorCOO>* out) {
*out = ::pyarrow_unwrap_sparse_tensor_coo(sparse_tensor);
Status unwrap_sparse_coo_tensor(PyObject* sparse_tensor,
std::shared_ptr<SparseCOOTensor>* out) {
*out = ::pyarrow_unwrap_sparse_coo_tensor(sparse_tensor);
if (*out) {
return Status::OK();
} else {
return Status::Invalid(
"Could not unwrap SparseTensorCOO from the passed Python object.");
"Could not unwrap SparseCOOTensor from the passed Python object.");
}
}

PyObject* wrap_sparse_tensor_coo(const std::shared_ptr<SparseTensorCOO>& sparse_tensor) {
return ::pyarrow_wrap_sparse_tensor_coo(sparse_tensor);
PyObject* wrap_sparse_coo_tensor(const std::shared_ptr<SparseCOOTensor>& sparse_tensor) {
return ::pyarrow_wrap_sparse_coo_tensor(sparse_tensor);
}

bool is_table(PyObject* table) { return ::pyarrow_is_table(table) != 0; }
Expand Down
12 changes: 6 additions & 6 deletions cpp/src/arrow/python/pyarrow.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,15 +75,15 @@ ARROW_PYTHON_EXPORT PyObject* wrap_tensor(const std::shared_ptr<Tensor>& tensor)

ARROW_PYTHON_EXPORT bool is_sparse_tensor_coo(PyObject* sparse_tensor);
ARROW_PYTHON_EXPORT Status
unwrap_sparse_tensor_coo(PyObject* sparse_tensor, std::shared_ptr<SparseTensorCOO>* out);
ARROW_PYTHON_EXPORT PyObject* wrap_sparse_tensor_coo(
const std::shared_ptr<SparseTensorCOO>& sparse_tensor);
unwrap_sparse_coo_tensor(PyObject* sparse_tensor, std::shared_ptr<SparseCOOTensor>* out);
ARROW_PYTHON_EXPORT PyObject* wrap_sparse_coo_tensor(
const std::shared_ptr<SparseCOOTensor>& sparse_tensor);

ARROW_PYTHON_EXPORT bool is_sparse_tensor_csr(PyObject* sparse_tensor);
ARROW_PYTHON_EXPORT Status
unwrap_sparse_tensor_csr(PyObject* sparse_tensor, std::shared_ptr<SparseTensorCSR>* out);
ARROW_PYTHON_EXPORT PyObject* wrap_sparse_tensor_csr(
const std::shared_ptr<SparseTensorCSR>& sparse_tensor);
unwrap_sparse_csr_matrix(PyObject* sparse_tensor, std::shared_ptr<SparseCSRMatrix>* out);
ARROW_PYTHON_EXPORT PyObject* wrap_sparse_csr_matrix(
const std::shared_ptr<SparseCSRMatrix>& sparse_tensor);

ARROW_PYTHON_EXPORT bool is_table(PyObject* table);
ARROW_PYTHON_EXPORT Status unwrap_table(PyObject* table, std::shared_ptr<Table>* out);
Expand Down
Loading