Skip to content

PERF: optimize DataFrame.sparse.from_spmatrix performance #32825

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

Merged
merged 18 commits into from
Mar 22, 2020
Merged

PERF: optimize DataFrame.sparse.from_spmatrix performance #32825

Show file tree
Hide file tree
Changes from 5 commits
Commits
Show all changes
18 commits
Select commit Hold shift + click to select a range
3ffe7f2
PERF: fix SparseArray._simple_new object initialization
jorisvandenbossche Mar 19, 2020
93bf825
Initial implementation
rth Mar 19, 2020
e21701b
Merge remote-tracking branch 'upstream/master' into opt-sparse-init
rth Mar 19, 2020
e095e7f
Improve docstring
rth Mar 19, 2020
11afe40
Add what's new
rth Mar 19, 2020
eda0732
Update doc/source/whatsnew/v1.1.0.rst
rth Mar 19, 2020
40f4cd6
Improve what's new
rth Mar 19, 2020
508fda5
Add random state
rth Mar 19, 2020
0053817
Add inline comment about sort_indices
rth Mar 19, 2020
42792a3
Use absolute import
rth Mar 19, 2020
c8f7abc
Update pandas/_libs/sparse.pyx
rth Mar 19, 2020
e541b0d
Rename variable
rth Mar 19, 2020
0a0bafc
Merge remote-tracking branch 'upstream/master' into opt-sparse-init
rth Mar 20, 2020
ce6619e
Use DataFrame._from_arrays(..., verify_integrity=False)
rth Mar 20, 2020
e063c8a
checkout upstream/master -- asv_bench/benchmarks/sparse.py
rth Mar 20, 2020
8efe3ad
Merge remote-tracking branch 'upstream/master' into opt-sparse-init
rth Mar 22, 2020
20c3685
Link to PR's by Joris in what's new
rth Mar 22, 2020
d750eb4
Use ensure_index when the columns/index is provided by the user
rth Mar 22, 2020
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
3 changes: 3 additions & 0 deletions doc/source/whatsnew/v1.1.0.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -224,6 +224,9 @@ Performance improvements
- The internal index method :meth:`~Index._shallow_copy` now copies cached attributes over to the new index,
avoiding creating these again on the new index. This can speed up many operations that depend on creating copies of
existing indexes (:issue:`28584`, :issue:`32640`, :issue:`32669`)
- Performance improvement when creating sparse :class:`DataFrame` from
``scipy.sparse`` matrices using the :meth:`DataFrame.sparse.from_spmatrix`
constructor (:issue:`32196`).
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

add the issue number from joris PR as well

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added PRs by Joris, there have been 5 PRs on this in total.

Ideally another what's new entry should be added since PR's by @jorisvandenbossche also made initialization of extension arrays faster under some conditions, as far as I understand. Though I would rather not add it here, nor am I competent on accurately formulating it.


.. ---------------------------------------------------------------------------

Expand Down
7 changes: 5 additions & 2 deletions pandas/_libs/sparse.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,18 +34,21 @@ cdef class IntIndex(SparseIndex):
length : integer
indices : array-like
Contains integers corresponding to the indices.
check_integrity : bool, default=True
Check integrity of the input.
"""

cdef readonly:
Py_ssize_t length, npoints
ndarray indices

def __init__(self, Py_ssize_t length, indices):
def __init__(self, Py_ssize_t length, indices, check_integrity=True):
self.length = length
self.indices = np.ascontiguousarray(indices, dtype=np.int32)
self.npoints = len(self.indices)

self.check_integrity()
if check_integrity:
self.check_integrity()

def __reduce__(self):
args = (self.length, self.indices)
Expand Down
21 changes: 15 additions & 6 deletions pandas/core/arrays/sparse/accessor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -227,15 +227,24 @@ def from_spmatrix(cls, data, index=None, columns=None):
1 0.0 1.0 0.0
2 0.0 0.0 1.0
"""
from pandas import DataFrame
from pandas import DataFrame, SparseDtype
from . import IntIndex, SparseArray

data = data.tocsc()
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could be tocsc(copy=False) for newer scipy versions, but the gain in performance is minimal anyway with respect to the total runtime.

index, columns = cls._prep_index(data, index, columns)
Copy link
Member

@jorisvandenbossche jorisvandenbossche Mar 22, 2020 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the problem is that this _pre_index doesn't actually return Index objects when columns or index is not None (and passing actual Index objects is required when doing verify_integrity=False)

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OK, I see. Thanks for confirming that passing duplicate columns names in an Index object is expected to work in _from_arrays. Will look into it. Thanks Joris!

Copy link
Member

@jorisvandenbossche jorisvandenbossche Mar 22, 2020 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That seems to be it, as doing pd.DataFrame.sparse.from_spmatrix(mat, columns=pd.Index(['a', 'a'])) instead of pd.DataFrame.sparse.from_spmatrix(mat, columns=['a', 'a']) works.

You can add in here a ensure_index call on both index and columns. (from pandas.core.indexes.api import ensure_index)

sparrays = [SparseArray.from_spmatrix(data[:, i]) for i in range(data.shape[1])]
data = dict(enumerate(sparrays))
result = DataFrame(data, index=index)
result.columns = columns
return result
n_rows, n_columns = data.shape
data.sort_indices()
Copy link
Contributor Author

@rth rth Mar 19, 2020 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It might be already done in tocsc, but that's a scipy implementation detail, and it doesn't really cost much. We need to make sure indices are sorted, since we create IntIndex with check_integrity=False that used to check for this.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe add a comment about that inline?

indices = data.indices
indptr = data.indptr
data = data.data
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you assign to a different variable? This in theory could cause mypy complaints (though ndarray resolving to Any at the moment probably allows it to pass)

dtype = SparseDtype(data.dtype, 0)
arrays = []
for i in range(n_columns):
sl = slice(indptr[i], indptr[i + 1])
idx = IntIndex(n_rows, indices[sl], check_integrity=False)
arr = SparseArray._simple_new(data[sl], idx, dtype)
arrays.append(arr)
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

FWIW, also tried with a generator here to avoid pre-allocating all the arrays, but it doesn't really matter. Most of the remaining run time is in DataFrame._from_arrays.

return DataFrame._from_arrays(arrays, columns=columns, index=index)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This line will be able to use the verify_integrity=False from #32858 (or if this PR goes in first, I can add it in that PR)


def to_dense(self):
"""
Expand Down