def test_tsvd_fit_transform(datatype, name, use_handle):
if name == 'blobs':
X, y = make_blobs(n_samples=500000, n_features=1000, random_state=0)
elif name == 'random':
pytest.skip('fails when using random dataset '
'used by sklearn for testing')
shape = 5000, 100
rng = check_random_state(42)
X = rng.randint(-100, 20, np.product(shape)).reshape(shape)
else:
n, p = 500, 5
rng = np.random.RandomState(0)
X = rng.randn(n, p) * .1 + np.array([3, 4, 2, 3, 5])
if name != 'blobs':
skpca = skTSVD(n_components=1)
Xsktsvd = skpca.fit_transform(X)
handle, stream = get_handle(use_handle)
cutsvd = cuTSVD(n_components=1, handle=handle)
Xcutsvd = cutsvd.fit_transform(X)
cutsvd.handle.sync()
if name != 'blobs':
assert array_equal(Xcutsvd, Xsktsvd, 1e-3, with_sign=True)
def test_dbscan_propagation(datatype, use_handle, out_dtype, n_samples):
X, y = make_blobs(n_samples,
centers=1,
cluster_std=8.0,
center_box=(-100.0, 100.0),
random_state=8)
X = X.astype(datatype)
handle, stream = get_handle(use_handle)
eps = 0.5
cuml_dbscan = cuDBSCAN(handle=handle,
eps=eps,
min_samples=5,
output_type='numpy')
cu_labels = cuml_dbscan.fit_predict(X, out_dtype=out_dtype)
sk_dbscan = skDBSCAN(eps=eps, min_samples=5)
sk_labels = sk_dbscan.fit_predict(X)
# Check the core points are equal
assert array_equal(cuml_dbscan.core_sample_indices_,
sk_dbscan.core_sample_indices_)
# Check the labels are correct
assert_dbscan_equal(sk_labels, cu_labels, X,
cuml_dbscan.core_sample_indices_, eps)
def test_rf_regression(special_reg, datatype, max_features, max_samples,
n_bins):
use_handle = True
X, y = special_reg
X = X.astype(datatype)
y = y.astype(datatype)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
train_size=0.8,
random_state=0)
# Create a handle for the cuml model
handle, stream = get_handle(use_handle, n_streams=1)
# Initialize and fit using cuML's random forest regression model
cuml_model = curfr(
max_features=max_features,
max_samples=max_samples,
n_bins=n_bins,
split_criterion=2,
min_samples_leaf=2,
random_state=123,
n_streams=1,
n_estimators=50,
handle=handle,
max_leaves=-1,
max_depth=16,
accuracy_metric="mse",
)
cuml_model.fit(X_train, y_train)
# predict using FIL
fil_preds = cuml_model.predict(X_test, predict_model="GPU")
cu_preds = cuml_model.predict(X_test, predict_model="CPU")
fil_preds = np.reshape(fil_preds, np.shape(cu_preds))
cu_r2 = r2_score(y_test, cu_preds, convert_dtype=datatype)
fil_r2 = r2_score(y_test, fil_preds, convert_dtype=datatype)
# Initialize, fit and predict using
# sklearn's random forest regression model
if X.shape[0] < 1000: # mode != "stress"
sk_model = skrfr(
n_estimators=50,
max_depth=16,
min_samples_split=2,
max_features=max_features,
random_state=10,
)
sk_model.fit(X_train, y_train)
sk_preds = sk_model.predict(X_test)
sk_r2 = r2_score(y_test, sk_preds, convert_dtype=datatype)
assert fil_r2 >= (sk_r2 - 0.07)
assert fil_r2 >= (cu_r2 - 0.02)
def test_rf_classification(small_clf, datatype, max_samples, max_features):
use_handle = True
X, y = small_clf
X = X.astype(datatype)
y = y.astype(np.int32)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
train_size=0.8,
random_state=0)
# Create a handle for the cuml model
handle, stream = get_handle(use_handle, n_streams=1)
# Initialize, fit and predict using cuML's
# random forest classification model
cuml_model = curfc(
max_features=max_features,
max_samples=max_samples,
n_bins=16,
split_criterion=0,
min_samples_leaf=2,
random_state=123,
n_streams=1,
n_estimators=40,
handle=handle,
max_leaves=-1,
max_depth=16,
)
cuml_model.fit(X_train, y_train)
fil_preds = cuml_model.predict(X_test,
predict_model="GPU",
threshold=0.5,
algo="auto")
cu_preds = cuml_model.predict(X_test, predict_model="CPU")
fil_preds = np.reshape(fil_preds, np.shape(cu_preds))
cuml_acc = accuracy_score(y_test, cu_preds)
fil_acc = accuracy_score(y_test, fil_preds)
if X.shape[0] < 500000:
sk_model = skrfc(
n_estimators=40,
max_depth=16,
min_samples_split=2,
max_features=max_features,
random_state=10,
)
sk_model.fit(X_train, y_train)
sk_preds = sk_model.predict(X_test)
sk_acc = accuracy_score(y_test, sk_preds)
assert fil_acc >= (sk_acc - 0.07)
assert fil_acc >= (
cuml_acc - 0.07
) # to be changed to 0.02. see issue #3910: https://github.com/rapidsai/cuml/issues/3910 # noqa
def test_rf_get_text(n_estimators, detailed_text):
X, y = make_classification(
n_samples=500,
n_features=10,
n_clusters_per_class=1,
n_informative=5,
random_state=94929,
n_classes=2,
)
X = X.astype(np.float32)
y = y.astype(np.int32)
# Create a handle for the cuml model
handle, stream = get_handle(True, n_streams=1)
# Initialize cuML Random Forest classification model
cuml_model = curfc(
handle=handle,
max_features=1.0,
max_samples=1.0,
n_bins=16,
split_criterion=0,
min_samples_leaf=2,
random_state=23707,
n_streams=1,
n_estimators=n_estimators,
max_leaves=-1,
max_depth=16,
)
# Train model on the data
cuml_model.fit(X, y)
if detailed_text:
text_output = cuml_model.get_detailed_text()
else:
text_output = cuml_model.get_summary_text()
# Test 1: Output is non-zero
assert "" != text_output
# Count the number of trees printed
tree_count = 0
for line in text_output.split("\n"):
if line.strip().startswith("Tree #"):
tree_count += 1
# Test 2: Correct number of trees are printed
assert n_estimators == tree_count
def test_dbscan(datatype, use_handle, nrows, ncols, max_mbytes_per_batch,
out_dtype):
if nrows == 500000 and pytest.max_gpu_memory < 32:
if pytest.adapt_stress_test:
nrows = nrows * pytest.max_gpu_memory // 32
else:
pytest.skip("Insufficient GPU memory for this test. "
"Re-run with 'CUML_ADAPT_STRESS_TESTS=True'")
n_samples = nrows
n_feats = ncols
X, y = make_blobs(n_samples=n_samples,
cluster_std=0.01,
n_features=n_feats,
random_state=0)
handle, stream = get_handle(use_handle)
eps = 1
cuml_dbscan = cuDBSCAN(handle=handle,
eps=eps,
min_samples=2,
max_mbytes_per_batch=max_mbytes_per_batch,
output_type='numpy')
cu_labels = cuml_dbscan.fit_predict(X, out_dtype=out_dtype)
if nrows < 500000:
sk_dbscan = skDBSCAN(eps=1, min_samples=2, algorithm="brute")
sk_labels = sk_dbscan.fit_predict(X)
# Check the core points are equal
assert array_equal(cuml_dbscan.core_sample_indices_,
sk_dbscan.core_sample_indices_)
# Check the labels are correct
assert_dbscan_equal(sk_labels, cu_labels, X,
cuml_dbscan.core_sample_indices_, eps)
if out_dtype == "int32" or out_dtype == np.int32:
assert cu_labels.dtype == np.int32
elif out_dtype == "int64" or out_dtype == np.int64:
assert cu_labels.dtype == np.int64
def test_rf_memory_leakage(small_clf, datatype, fil_sparse_format, n_iter):
use_handle = True
X, y = small_clf
X = X.astype(datatype)
y = y.astype(np.int32)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
train_size=0.8,
random_state=0)
# Create a handle for the cuml model
handle, stream = get_handle(use_handle, n_streams=1)
# Warmup. Some modules that are used in RF allocate space on the device
# and consume memory. This is to make sure that the allocation is done
# before the first call to get_memory_info.
base_model = curfc(handle=handle)
base_model.fit(X_train, y_train)
handle.sync() # just to be sure
free_mem = cuda.current_context().get_memory_info()[0]
def test_for_memory_leak():
cuml_mods = curfc(handle=handle)
cuml_mods.fit(X_train, y_train)
handle.sync() # just to be sure
# Calculate the memory free after fitting the cuML model
delta_mem = free_mem - cuda.current_context().get_memory_info()[0]
assert delta_mem == 0
for i in range(2):
cuml_mods.predict(
X_test,
predict_model="GPU",
fil_sparse_format=fil_sparse_format,
)
handle.sync() # just to be sure
# Calculate the memory free after predicting the cuML model
delta_mem = free_mem - cuda.current_context().get_memory_info()[0]
assert delta_mem == 0
for i in range(n_iter):
test_for_memory_leak()
def test_pca_inverse_transform(datatype, input_type, name, use_handle, nrows):
if name == 'blobs':
pytest.skip('fails when using blobs dataset')
X, y = make_blobs(n_samples=500000, n_features=1000, random_state=0)
else:
rng = np.random.RandomState(0)
n, p = nrows, 3
X = rng.randn(n, p) # spherical data
X[:, 1] *= .00001 # make middle component relatively small
X += [3, 4, 2] # make a large mean
handle, stream = get_handle(use_handle)
cupca = cuPCA(n_components=2, handle=handle)
X_cupca = cupca.fit_transform(X)
input_gdf = cupca.inverse_transform(X_cupca)
cupca.handle.sync()
assert array_equal(input_gdf, X, 5e-5, with_sign=True)
def test_pca_fit_then_transform(datatype, input_type, name, use_handle):
blobs_n_samples = 500000
if name == 'blobs' and pytest.max_gpu_memory < 32:
if pytest.adapt_stress_test:
blobs_n_samples = int(blobs_n_samples * pytest.max_gpu_memory / 32)
else:
pytest.skip("Insufficient GPU memory for this test."
"Re-run with 'CUML_ADAPT_STRESS_TESTS=True'")
if name == 'blobs':
X, y = make_blobs(n_samples=blobs_n_samples,
n_features=1000,
random_state=0)
elif name == 'iris':
iris = datasets.load_iris()
X = iris.data
else:
X, Y = make_multilabel_classification(n_samples=500,
n_classes=2,
n_labels=1,
allow_unlabeled=False,
random_state=1)
if name != 'blobs':
skpca = skPCA(n_components=2)
skpca.fit(X)
Xskpca = skpca.transform(X)
handle, stream = get_handle(use_handle)
cupca = cuPCA(n_components=2, handle=handle)
cupca.fit(X)
X_cupca = cupca.transform(X)
cupca.handle.sync()
if name != 'blobs':
assert array_equal(X_cupca, Xskpca, 1e-3, with_sign=False)
assert Xskpca.shape[0] == X_cupca.shape[0]
assert Xskpca.shape[1] == X_cupca.shape[1]
def test_tsvd_fit(datatype, name, use_handle):
if name == 'blobs':
X, y = make_blobs(n_samples=500000, n_features=1000, random_state=0)
elif name == 'random':
pytest.skip('fails when using random dataset '
'used by sklearn for testing')
shape = 5000, 100
rng = check_random_state(42)
X = rng.randint(-100, 20, np.product(shape)).reshape(shape)
else:
n, p = 500, 5
rng = np.random.RandomState(0)
X = rng.randn(n, p) * .1 + np.array([3, 4, 2, 3, 5])
if name != 'blobs':
sktsvd = skTSVD(n_components=1)
sktsvd.fit(X)
handle, stream = get_handle(use_handle)
cutsvd = cuTSVD(n_components=1, handle=handle)
cutsvd.fit(X)
cutsvd.handle.sync()
if name != 'blobs':
for attr in [
'singular_values_', 'components_', 'explained_variance_ratio_'
]:
with_sign = False if attr in ['components_'] else True
assert array_equal(getattr(cutsvd, attr),
getattr(sktsvd, attr),
0.4,
with_sign=with_sign)
def test_pca_fit(datatype, input_type, name, use_handle):
if name == 'blobs':
pytest.skip('fails when using blobs dataset')
X, y = make_blobs(n_samples=500000, n_features=1000, random_state=0)
elif name == 'digits':
X, _ = datasets.load_digits(return_X_y=True)
else:
X, Y = make_multilabel_classification(n_samples=500,
n_classes=2,
n_labels=1,
allow_unlabeled=False,
random_state=1)
skpca = skPCA(n_components=2)
skpca.fit(X)
handle, stream = get_handle(use_handle)
cupca = cuPCA(n_components=2, handle=handle)
cupca.fit(X)
cupca.handle.sync()
for attr in [
'singular_values_', 'components_', 'explained_variance_',
'explained_variance_ratio_'
]:
with_sign = False if attr in ['components_'] else True
print(attr)
print(getattr(cupca, attr))
print(getattr(skpca, attr))
cuml_res = (getattr(cupca, attr))
skl_res = getattr(skpca, attr)
assert array_equal(cuml_res, skl_res, 1e-3, with_sign=with_sign)
def test_rf_regression_sparse(special_reg, datatype, fil_sparse_format, algo):
use_handle = True
num_treees = 50
X, y = special_reg
X = X.astype(datatype)
y = y.astype(datatype)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
train_size=0.8,
random_state=0)
# Create a handle for the cuml model
handle, stream = get_handle(use_handle, n_streams=1)
# Initialize and fit using cuML's random forest regression model
cuml_model = curfr(
n_bins=16,
split_criterion=2,
min_samples_leaf=2,
random_state=123,
n_streams=1,
n_estimators=num_treees,
handle=handle,
max_leaves=-1,
max_depth=40,
accuracy_metric="mse",
)
cuml_model.fit(X_train, y_train)
# predict using FIL
if (not fil_sparse_format or algo == "tree_reorg" or algo
== "batch_tree_reorg") or fil_sparse_format == "not_supported":
with pytest.raises(ValueError):
fil_preds = cuml_model.predict(
X_test,
predict_model="GPU",
fil_sparse_format=fil_sparse_format,
algo=algo,
)
else:
fil_preds = cuml_model.predict(
X_test,
predict_model="GPU",
fil_sparse_format=fil_sparse_format,
algo=algo,
)
fil_preds = np.reshape(fil_preds, np.shape(y_test))
fil_r2 = r2_score(y_test, fil_preds, convert_dtype=datatype)
fil_model = cuml_model.convert_to_fil_model()
with cuml.using_output_type("numpy"):
fil_model_preds = fil_model.predict(X_test)
fil_model_preds = np.reshape(fil_model_preds, np.shape(y_test))
fil_model_r2 = r2_score(y_test,
fil_model_preds,
convert_dtype=datatype)
assert fil_r2 == fil_model_r2
tl_model = cuml_model.convert_to_treelite_model()
assert num_treees == tl_model.num_trees
assert X.shape[1] == tl_model.num_features
# Initialize, fit and predict using
# sklearn's random forest regression model
if X.shape[0] < 1000: # mode != "stress":
sk_model = skrfr(
n_estimators=50,
max_depth=40,
min_samples_split=2,
random_state=10,
)
sk_model.fit(X_train, y_train)
sk_preds = sk_model.predict(X_test)
sk_r2 = r2_score(y_test, sk_preds, convert_dtype=datatype)
assert fil_r2 >= (sk_r2 - 0.07)
def test_rf_classification_sparse(small_clf, datatype, fil_sparse_format,
algo):
use_handle = True
num_treees = 50
X, y = small_clf
X = X.astype(datatype)
y = y.astype(np.int32)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
train_size=0.8,
random_state=0)
# Create a handle for the cuml model
handle, stream = get_handle(use_handle, n_streams=1)
# Initialize, fit and predict using cuML's
# random forest classification model
cuml_model = curfc(
n_bins=16,
split_criterion=0,
min_samples_leaf=2,
random_state=123,
n_streams=1,
n_estimators=num_treees,
handle=handle,
max_leaves=-1,
max_depth=40,
)
cuml_model.fit(X_train, y_train)
if (not fil_sparse_format or algo == "tree_reorg" or algo
== "batch_tree_reorg") or fil_sparse_format == "not_supported":
with pytest.raises(ValueError):
fil_preds = cuml_model.predict(
X_test,
predict_model="GPU",
threshold=0.5,
fil_sparse_format=fil_sparse_format,
algo=algo,
)
else:
fil_preds = cuml_model.predict(
X_test,
predict_model="GPU",
threshold=0.5,
fil_sparse_format=fil_sparse_format,
algo=algo,
)
fil_preds = np.reshape(fil_preds, np.shape(y_test))
fil_acc = accuracy_score(y_test, fil_preds)
np.testing.assert_almost_equal(fil_acc,
cuml_model.score(X_test, y_test))
fil_model = cuml_model.convert_to_fil_model()
with cuml.using_output_type("numpy"):
fil_model_preds = fil_model.predict(X_test)
fil_model_acc = accuracy_score(y_test, fil_model_preds)
assert fil_acc == fil_model_acc
tl_model = cuml_model.convert_to_treelite_model()
assert num_treees == tl_model.num_trees
assert X.shape[1] == tl_model.num_features
if X.shape[0] < 500000:
sk_model = skrfc(
n_estimators=50,
max_depth=40,
min_samples_split=2,
random_state=10,
)
sk_model.fit(X_train, y_train)
sk_preds = sk_model.predict(X_test)
sk_acc = accuracy_score(y_test, sk_preds)
assert fil_acc >= (sk_acc - 0.07)
def rf_classification(datatype, array_type, max_features, max_samples,
fixture):
X, y = fixture
X = X.astype(datatype[0])
y = y.astype(np.int32)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
train_size=0.8,
random_state=0)
X_test = X_test.astype(datatype[1])
n_streams = 1
handle, stream = get_handle(True, n_streams=n_streams)
# Initialize, fit and predict using cuML's
# random forest classification model
cuml_model = curfc(
max_features=max_features,
max_samples=max_samples,
n_bins=16,
split_criterion=0,
min_samples_leaf=2,
random_state=123,
n_estimators=40,
handle=handle,
max_leaves=-1,
max_depth=16,
n_streams=n_streams,
)
if array_type == "dataframe":
X_train_df = cudf.DataFrame(X_train)
y_train_df = cudf.Series(y_train)
X_test_df = cudf.DataFrame(X_test)
cuml_model.fit(X_train_df, y_train_df)
cu_proba_gpu = cuml_model.predict_proba(X_test_df).to_numpy()
cu_preds_cpu = cuml_model.predict(X_test_df,
predict_model="CPU").to_numpy()
cu_preds_gpu = cuml_model.predict(X_test_df,
predict_model="GPU").to_numpy()
else:
cuml_model.fit(X_train, y_train)
cu_proba_gpu = cuml_model.predict_proba(X_test)
cu_preds_cpu = cuml_model.predict(X_test, predict_model="CPU")
cu_preds_gpu = cuml_model.predict(X_test, predict_model="GPU")
np.testing.assert_array_equal(cu_preds_gpu, np.argmax(cu_proba_gpu,
axis=1))
cu_acc_cpu = accuracy_score(y_test, cu_preds_cpu)
cu_acc_gpu = accuracy_score(y_test, cu_preds_gpu)
assert cu_acc_cpu == pytest.approx(cu_acc_gpu, abs=0.01, rel=0.1)
# sklearn random forest classification model
# initialization, fit and predict
if y.size < 500000:
sk_model = skrfc(
n_estimators=40,
max_depth=16,
min_samples_split=2,
max_features=max_features,
random_state=10,
)
sk_model.fit(X_train, y_train)
sk_preds = sk_model.predict(X_test)
sk_acc = accuracy_score(y_test, sk_preds)
sk_proba = sk_model.predict_proba(X_test)
assert cu_acc_cpu >= sk_acc - 0.07
assert cu_acc_gpu >= sk_acc - 0.07
# 0.06 is the highest relative error observed on CI, within
# 0.0061 absolute error boundaries seen previously
check_predict_proba(cu_proba_gpu, sk_proba, y_test, 0.1)
