Replace direct np.random.* calls with np.random.RandomState instances

monai/data/synthetic.py

@@ -62,7 +62,7 @@ def create_test_image_2d(
         raise ValueError(f"the minimal size {min_size} of the image should be larger than `2 * rad_max` 2x{rad_max}.")
 
     image = np.zeros((height, width))
-    rs: np.random.RandomState = np.random.random.__self__ if random_state is None else random_state  # type: ignore
+    rs: np.random.RandomState = np.random.RandomState() if random_state is None else random_state
 
     for _ in range(num_objs):
         x = rs.randint(rad_max, height - rad_max)
@@ -139,7 +139,7 @@ def create_test_image_3d(
         raise ValueError(f"the minimal size {min_size} of the image should be larger than `2 * rad_max` 2x{rad_max}.")
 
     image = np.zeros((height, width, depth))
-    rs: np.random.RandomState = np.random.random.__self__ if random_state is None else random_state  # type: ignore
+    rs: np.random.RandomState = np.random.RandomState() if random_state is None else random_state
 
     for _ in range(num_objs):
         x = rs.randint(rad_max, height - rad_max)

monai/data/utils.py

@@ -120,7 +120,7 @@ def get_random_patch(
     """
 
     # choose the minimal corner of the patch
-    rand_int = np.random.randint if rand_state is None else rand_state.randint
+    rand_int = np.random.RandomState().randint if rand_state is None else rand_state.randint
     min_corner = tuple(rand_int(0, ms - ps + 1) if ms > ps else 0 for ms, ps in zip(dims, patch_size))
 
     # create the slices for each dimension which define the patch in the source array

monai/transforms/signal/array.py

@@ -273,7 +273,7 @@ def __call__(self, signal: NdarrayOrTensor) -> NdarrayOrTensor:
         data = convert_to_tensor(self.freqs * time_partial)
         sine_partial = self.magnitude * torch.sin(data)
 
-        loc = np.random.choice(range(length))
+        loc = self.R.choice(range(length))
         signal = paste(signal, sine_partial, (loc,))
 
         return signal
@@ -354,7 +354,7 @@ def __call__(self, signal: NdarrayOrTensor) -> NdarrayOrTensor:
         time_partial = np.arange(0, round(self.fracs * length), 1)
         squaredpulse_partial = self.magnitude * squarepulse(self.freqs * time_partial)
 
-        loc = np.random.choice(range(length))
+        loc = self.R.choice(range(length))
         signal = paste(signal, squaredpulse_partial, (loc,))
 
         return signal

monai/transforms/utils.py

@@ -666,7 +666,7 @@ def generate_pos_neg_label_crop_centers(
 
     """
     if rand_state is None:
-        rand_state = np.random.random.__self__  # type: ignore
+        rand_state = np.random.RandomState()
 
     centers = []
     fg_indices = np.asarray(fg_indices) if isinstance(fg_indices, Sequence) else fg_indices
@@ -721,7 +721,7 @@ def generate_label_classes_crop_centers(
 
     """
     if rand_state is None:
-        rand_state = np.random.random.__self__  # type: ignore
+        rand_state = np.random.RandomState()
 
     if num_samples < 1:
         raise ValueError(f"num_samples must be an int number and greater than 0, got {num_samples}.")
@@ -1585,7 +1585,7 @@ def get_extreme_points(
     """
     check_non_lazy_pending_ops(img, name="get_extreme_points")
     if rand_state is None:
-        rand_state = np.random.random.__self__  # type: ignore
+        rand_state = np.random.RandomState()
     indices = where(img != background)
     if np.size(indices[0]) == 0:
         raise ValueError("get_extreme_points: no foreground object in mask!")

monai/utils/ordering.py

@@ -202,6 +202,6 @@ def random_idx(rows: int, cols: int, depths: int | None = None) -> np.ndarray:
                     idx.append((r, c))
 
         idx_np = np.array(idx)
-        np.random.shuffle(idx_np)
+        np.random.RandomState().shuffle(idx_np)
 
         return idx_np