Fix numpy 2.0 deprecations (#5866)

Split from #5799 * `product` -> `prod` * Coercing single element arrays to it's scalar value via `int` or similar appears deprecated, so adjusted not to do this where possible Authors: - Matthew Roeschke (https://github.com/mroeschke) Approvers: - Dante Gama Dessavre (https://github.com/dantegd) URL: #5866
rapidsai · Apr 25, 2024 · d3d0ff3 · d3d0ff3
1 parent 4394a2e
commit d3d0ff3
Show file tree

Hide file tree

Showing 8 changed files with 27 additions and 40 deletions.
diff --git a/python/cuml/datasets/blobs.py b/python/cuml/datasets/blobs.py
@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2020-2023, NVIDIA CORPORATION.
+# Copyright (c) 2020-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -209,12 +209,8 @@ def make_blobs(
         proba_samples_per_center = np.array(n_samples_per_center) / np.sum(
             n_samples_per_center
         )
-        np_seed = int(generator.randint(n_samples, size=1))
-        np.random.seed(np_seed)
-        shuffled_sample_indices = cp.array(
-            np.random.choice(
-                n_centers, n_samples, replace=True, p=proba_samples_per_center
-            )
+        shuffled_sample_indices = generator.choice(
+            n_centers, n_samples, replace=True, p=proba_samples_per_center
         )
         for i, (n, std) in enumerate(zip(n_samples_per_center, cluster_std)):
             center_indices = cp.where(shuffled_sample_indices == i)

diff --git a/python/cuml/datasets/classification.py b/python/cuml/datasets/classification.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2023, NVIDIA CORPORATION.
+# Copyright (c) 2020-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -120,7 +120,7 @@ def make_classification(
         [-0.8817671  -0.84549576  0.1845096   0.02556021]]
 
         >>> print(y)
-        [0 1 0 1 1 0 0 1 0 0]
+        [1 0 1 1 1 1 1 1 1 0]
 
     Parameters
     ----------
@@ -229,8 +229,6 @@ def make_classification(
     cuml.internals.set_api_output_type("cupy")
 
     generator = _create_rs_generator(random_state)
-    np_seed = int(generator.randint(n_samples, size=1))
-    np.random.seed(np_seed)
 
     # Count features, clusters and samples
     if n_informative + n_redundant + n_repeated > n_features:
@@ -307,13 +305,8 @@ def make_classification(
         proba_samples_per_cluster = np.array(n_samples_per_cluster) / np.sum(
             n_samples_per_cluster
         )
-        shuffled_sample_indices = cp.array(
-            np.random.choice(
-                n_clusters,
-                n_samples,
-                replace=True,
-                p=proba_samples_per_cluster,
-            )
+        shuffled_sample_indices = generator.choice(
+            n_clusters, n_samples, replace=True, p=proba_samples_per_cluster
         )
         for k, centroid in enumerate(centroids):
             centroid_indices = cp.where(shuffled_sample_indices == k)

diff --git a/python/cuml/internals/array.py b/python/cuml/internals/array.py
@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2020-2023, NVIDIA CORPORATION.
+# Copyright (c) 2020-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -447,7 +447,7 @@ def is_host_accessible(self):
     @cached_property
     def size(self):
         return (
-            host_xpy.product(self._array_interface["shape"])
+            host_xpy.prod(self._array_interface["shape"])
             * host_xpy.dtype(self._array_interface["typestr"]).itemsize
         )
 

diff --git a/python/cuml/linear_model/ridge.pyx b/python/cuml/linear_model/ridge.pyx
@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2019-2023, NVIDIA CORPORATION.
+# Copyright (c) 2019-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -101,7 +101,7 @@ class Ridge(UniversalBase,
         >>> from cuml import Ridge
         >>> from cuml.linear_model import Ridge
 
-        >>> alpha = cp.array([1e-5])
+        >>> alpha = 1e-5
         >>> ridge = Ridge(alpha=alpha, fit_intercept=True, normalize=False,
         ...               solver="eig")
 

diff --git a/python/cuml/multiclass/multiclass.py b/python/cuml/multiclass/multiclass.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2023, NVIDIA CORPORATION.
+# Copyright (c) 2020-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -63,7 +63,7 @@ class MulticlassClassifier(Base, ClassifierMixin):
         >>> cls.fit(X,y)
         MulticlassClassifier()
         >>> cls.predict(X)
-        array([2, 0, 2, 2, 2, 1, 1, 0, 1, 1])
+        array([1, 1, 1, 1, 1, 1, 2, 1, 1, 2])
 
     Parameters
     ----------
@@ -228,7 +228,7 @@ class OneVsRestClassifier(MulticlassClassifier):
         >>> cls.fit(X,y)
         OneVsRestClassifier()
         >>> cls.predict(X)
-        array([2, 0, 2, 2, 2, 1, 1, 0, 1, 1])
+        array([1, 1, 1, 1, 1, 1, 2, 1, 1, 2])
 
 
     Parameters
@@ -303,7 +303,7 @@ class OneVsOneClassifier(MulticlassClassifier):
         >>> cls.fit(X,y)
         OneVsOneClassifier()
         >>> cls.predict(X)
-        array([2, 0, 2, 2, 2, 1, 1, 0, 1, 1])
+        array([1, 1, 1, 1, 1, 1, 2, 1, 1, 2])
 
     Parameters
     ----------

diff --git a/python/cuml/neighbors/nearest_neighbors.pyx b/python/cuml/neighbors/nearest_neighbors.pyx
@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2019-2022, NVIDIA CORPORATION.
+# Copyright (c) 2019-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -253,11 +253,11 @@ class NearestNeighbors(UniversalBase,
         >>> # print results
         >>> print(indices)
         0  1  2
-        0  0  1  3
-        1  1  0  2
+        0  0  3  1
+        1  1  3  0
         2  2  4  0
-        3  3  0  2
-        4  4  2  3
+        3  3  0  1
+        4  4  2  0
         >>> print(distances) # doctest: +SKIP
                 0          1          2
         0  0.007812  24.786566  26.399996

diff --git a/python/cuml/tests/test_coordinate_descent.py b/python/cuml/tests/test_coordinate_descent.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2019-2023, NVIDIA CORPORATION.
+# Copyright (c) 2019-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -27,7 +27,6 @@
 
 
 @pytest.mark.parametrize("datatype", [np.float32, np.float64])
-@pytest.mark.parametrize("X_type", ["ndarray"])
 @pytest.mark.parametrize("alpha", [0.1, 0.001])
 @pytest.mark.parametrize("algorithm", ["cyclic", "random"])
 @pytest.mark.parametrize(
@@ -42,7 +41,7 @@
     ],
 )
 @pytest.mark.filterwarnings("ignore:Objective did not converge::sklearn[.*]")
-def test_lasso(datatype, X_type, alpha, algorithm, nrows, column_info):
+def test_lasso(datatype, alpha, algorithm, nrows, column_info):
     ncols, n_info = column_info
     X, y = make_regression(
         n_samples=nrows, n_features=ncols, n_informative=n_info, random_state=0
@@ -53,7 +52,7 @@ def test_lasso(datatype, X_type, alpha, algorithm, nrows, column_info):
         X, y, train_size=0.8, random_state=0
     )
     cu_lasso = cuLasso(
-        alpha=np.array([alpha]),
+        alpha=alpha,
         fit_intercept=True,
         max_iter=1000,
         selection=algorithm,
@@ -165,7 +164,6 @@ def test_weighted_cd(datatype, model, fit_intercept, distribution):
 
 
 @pytest.mark.parametrize("datatype", [np.float32, np.float64])
-@pytest.mark.parametrize("X_type", ["ndarray"])
 @pytest.mark.parametrize("alpha", [0.2, 0.7])
 @pytest.mark.parametrize("algorithm", ["cyclic", "random"])
 @pytest.mark.parametrize(
@@ -180,7 +178,7 @@ def test_weighted_cd(datatype, model, fit_intercept, distribution):
     ],
 )
 @pytest.mark.filterwarnings("ignore:Objective did not converge::sklearn[.*]")
-def test_elastic_net(datatype, X_type, alpha, algorithm, nrows, column_info):
+def test_elastic_net(datatype, alpha, algorithm, nrows, column_info):
     ncols, n_info = column_info
     X, y = make_regression(
         n_samples=nrows, n_features=ncols, n_informative=n_info, random_state=0
@@ -192,7 +190,7 @@ def test_elastic_net(datatype, X_type, alpha, algorithm, nrows, column_info):
     )
 
     elastic_cu = cuElasticNet(
-        alpha=np.array([alpha]),
+        alpha=alpha,
         fit_intercept=True,
         max_iter=1000,
         selection=algorithm,

diff --git a/python/cuml/tests/test_incremental_pca.py b/python/cuml/tests/test_incremental_pca.py
@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2020-2023, NVIDIA CORPORATION.
+# Copyright (c) 2020-2024, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -131,7 +131,7 @@ def test_partial_fit(
     sk_t = sk_ipca.transform(X)
     sk_inv = sk_ipca.inverse_transform(sk_t)
 
-    assert array_equal(cu_inv, sk_inv, 5e-5, with_sign=True)
+    assert array_equal(cu_inv, sk_inv, 6e-5, with_sign=True)
 
 
 def test_exceptions():