Skip to content

Speedup dask LogisticRegression tests #6607

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
rapids-bot merged 7 commits into from
Apr 30, 2025
Merged

Speedup dask LogisticRegression tests #6607

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
4b57c3e
Speedup `test_standardization_on_scaled_dataset`
jcrist Apr 29, 2025
aea6c25
Speedup `test_sparse_nlp20news`
jcrist Apr 29, 2025
755675d
Reduce `n_classes`
jcrist Apr 29, 2025
ad77d08
Remove more scaling parametrization
jcrist Apr 29, 2025
3784f97
_Do_ parametrize `test_n_classes` on `n_classes`
jcrist Apr 29, 2025
727eb6d
Don't parametrize some tests on `fit_intercept`
jcrist Apr 29, 2025
db855ef
Don't needlessly parametrize `test_elasticnet` by `datatype`
jcrist Apr 29, 2025
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
71 changes: 23 additions & 48 deletions python/cuml/cuml/tests/dask/test_dask_logistic_regression.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,6 @@
#

import random
from functools import partial

import cudf
import cupy as cp
Expand Down Expand Up @@ -427,31 +426,29 @@ def assert_small(X, y, n_classes):


@pytest.mark.parametrize("n_parts", [2, 23])
@pytest.mark.parametrize("fit_intercept", [False, True])
@pytest.mark.parametrize("n_classes", [8])
def test_n_classes(n_parts, fit_intercept, n_classes, client):
datatype = np.float32 if fit_intercept else np.float64
@pytest.mark.parametrize("n_classes", [2, 6])
def test_n_classes(n_parts, n_classes, client):
nrows = int(1e5) if n_classes < 5 else int(2e5)
lr = _test_lbfgs(
nrows=nrows,
ncols=20,
n_parts=n_parts,
fit_intercept=fit_intercept,
datatype=datatype,
fit_intercept=True,
Comment thread
betatim marked this conversation as resolved.
datatype=np.float32,
delayed=True,
client=client,
penalty="l2",
n_classes=n_classes,
)

assert lr._num_classes == n_classes
assert lr.dtype == datatype
assert lr.dtype == np.float32


@pytest.mark.mg
@pytest.mark.parametrize("fit_intercept", [False, True])
@pytest.mark.parametrize("delayed", [True])
@pytest.mark.parametrize("n_classes", [2, 8])
@pytest.mark.parametrize("n_classes", [2, 6])
@pytest.mark.parametrize("C", [1.0, 10.0])
def test_l1(fit_intercept, delayed, n_classes, C, client):
datatype = np.float64 if fit_intercept else np.float32
Expand All @@ -478,13 +475,10 @@ def test_l1(fit_intercept, delayed, n_classes, C, client):

@pytest.mark.mg
@pytest.mark.parametrize("fit_intercept", [False, True])
@pytest.mark.parametrize("datatype", [np.float32, np.float64])
@pytest.mark.parametrize("delayed", [True])
@pytest.mark.parametrize("n_classes", [2, 8])
@pytest.mark.parametrize("n_classes", [2, 6])
@pytest.mark.parametrize("l1_ratio", [0.2, 0.8])
def test_elasticnet(
fit_intercept, datatype, delayed, n_classes, l1_ratio, client
):
def test_elasticnet(fit_intercept, delayed, n_classes, l1_ratio, client):
datatype = np.float32 if fit_intercept else np.float64

nrows = int(1e5) if n_classes < 5 else int(2e5)
Expand All @@ -511,7 +505,6 @@ def test_elasticnet(


@pytest.mark.mg
@pytest.mark.parametrize("fit_intercept", [False, True])
@pytest.mark.parametrize(
"reg_dtype",
[
Expand All @@ -521,33 +514,27 @@ def test_elasticnet(
(("elasticnet", 2.0, 0.2), np.float64),
],
)
@pytest.mark.parametrize("n_classes", [2, 8])
def test_sparse_from_dense(fit_intercept, reg_dtype, n_classes, client):
def test_sparse_from_dense(reg_dtype, client):
penalty, C, l1_ratio = reg_dtype[0]
datatype = reg_dtype[1]

nrows = int(1e5) if n_classes < 5 else int(2e5)

_convert_index = np.int32 if random.choice([True, False]) else np.int64

run_test = partial(
_test_lbfgs,
nrows=nrows,
lr = _test_lbfgs(
nrows=int(1e5),
ncols=20,
n_parts=2,
fit_intercept=fit_intercept,
fit_intercept=True,
datatype=datatype,
delayed=True,
client=client,
penalty=penalty,
n_classes=n_classes,
n_classes=2,
C=C,
l1_ratio=l1_ratio,
convert_to_sparse=True,
_convert_index=_convert_index,
)

lr = run_test()
assert lr.dtype == datatype
assert lr.index_dtype == _convert_index

Expand Down Expand Up @@ -585,15 +572,14 @@ def test_sparse_nlp20news(dtype, nlp_20news, client):

from sklearn.linear_model import LogisticRegression as CPULR

cpu = CPULR(C=20.0)
cpu = CPULR(C=20.0, solver="saga")
cpu.fit(X_train, y_train)
cpu_preds = cpu.predict(X_test)
cpu_score = accuracy_score(y_test, cpu_preds.tolist())
assert cuml_score >= cpu_score or np.abs(cuml_score - cpu_score) < 1e-3


@pytest.mark.parametrize("fit_intercept", [False, True])
def test_exception_one_label(fit_intercept, client):
def test_exception_one_label(client):
n_parts = 2
datatype = "float32"

Expand All @@ -605,16 +591,10 @@ def test_exception_one_label(fit_intercept, client):

from cuml.dask.linear_model import LogisticRegression as cumlLBFGS_dask

mg = cumlLBFGS_dask(fit_intercept=fit_intercept, verbose=6)
mg = cumlLBFGS_dask(fit_intercept=True, verbose=6)
with pytest.raises(RuntimeError, match=err_msg):
mg.fit(X_df, y_df)

from sklearn.linear_model import LogisticRegression

lr = LogisticRegression(fit_intercept=fit_intercept)
with pytest.raises(ValueError, match=err_msg):
lr.fit(X, y)


@pytest.mark.mg
@pytest.mark.parametrize("fit_intercept", [False, True])
Expand All @@ -628,9 +608,8 @@ def test_exception_one_label(fit_intercept, client):
],
)
@pytest.mark.parametrize("delayed", [False])
@pytest.mark.parametrize("n_classes", [2, 8])
def test_standardization_on_normal_dataset(
fit_intercept, reg_dtype, delayed, n_classes, client
fit_intercept, reg_dtype, delayed, client
):

regularization = reg_dtype[0]
Expand All @@ -639,19 +618,17 @@ def test_standardization_on_normal_dataset(
C = regularization[1]
l1_ratio = regularization[2]

nrows = int(1e5) if n_classes < 5 else int(2e5)

# test correctness compared with scikit-learn
lr = _test_lbfgs(
nrows=nrows,
nrows=int(1e5),
ncols=20,
n_parts=2,
fit_intercept=fit_intercept,
datatype=datatype,
delayed=delayed,
client=client,
penalty=penalty,
n_classes=n_classes,
n_classes=2,
Comment thread
jcrist marked this conversation as resolved.
C=C,
l1_ratio=l1_ratio,
standardization=True,
Expand Down Expand Up @@ -712,20 +689,18 @@ def adjust_standardization_model_for_comparison(
],
)
@pytest.mark.parametrize("delayed", [False])
@pytest.mark.parametrize("ncol_and_nclasses", [(2, 2), (6, 4), (100, 10)])
def test_standardization_on_scaled_dataset(
fit_intercept, reg_dtype, delayed, ncol_and_nclasses, client
fit_intercept, reg_dtype, delayed, client
):

regularization = reg_dtype[0]
datatype = reg_dtype[1]

penalty = regularization[0]
C = regularization[1]
l1_ratio = regularization[2]
n_classes = ncol_and_nclasses[1]
nrows = int(1e5) if n_classes < 5 else int(2e5)
ncols = ncol_and_nclasses[0]
nrows = int(2e5)
ncols = 20
n_classes = 6
n_info = ncols
n_redundant = 0
n_parts = 2
Expand Down