Fix

Author: betatim

sklearn/metrics/_regression.py

@@ -1351,19 +1351,19 @@ def max_error(y_true, y_pred):
 
 def _mean_tweedie_deviance(y_true, y_pred, sample_weight, power):
     """Mean Tweedie deviance regression loss."""
-    xp, _ = get_namespace(y_true, y_pred)
+    xp, _, device_ = get_namespace_and_device(y_true, y_pred)
     p = power
-    zero = xp.asarray(0, dtype=y_true.dtype, device=y_true.device)
+    zero = xp.asarray(0, dtype=y_true.dtype, device=device_)
     if p < 0:
         # 'Extreme stable', y any real number, y_pred > 0
         dev = 2 * (
             xp.pow(
                 xp.where(y_true > 0, y_true, zero),
-                xp.asarray(2 - p, device=y_true.device),
+                xp.asarray(2 - p, device=device_),
             )
             / ((1 - p) * (2 - p))
-            - y_true * xp.pow(y_pred, xp.asarray(1 - p, device=y_true.device)) / (1 - p)
-            + xp.pow(y_pred, xp.asarray(2 - p, device=y_true.device)) / (2 - p)
+            - y_true * xp.pow(y_pred, xp.asarray(1 - p, device=device_)) / (1 - p)
+            + xp.pow(y_pred, xp.asarray(2 - p, device=device_)) / (2 - p)
         )
     elif p == 0:
         # Normal distribution, y and y_pred any real number
@@ -1376,10 +1376,9 @@ def _mean_tweedie_deviance(y_true, y_pred, sample_weight, power):
         dev = 2 * (xp.log(y_pred / y_true) + y_true / y_pred - 1)
     else:
         dev = 2 * (
-            xp.pow(y_true, xp.asarray(2 - p, device=y_true.device))
-            / ((1 - p) * (2 - p))
-            - y_true * xp.pow(y_pred, xp.asarray(1 - p, device=y_true.device)) / (1 - p)
-            + xp.pow(y_pred, xp.asarray(2 - p, device=y_true.device)) / (2 - p)
+            xp.pow(y_true, xp.asarray(2 - p, device=device_)) / ((1 - p) * (2 - p))
+            - y_true * xp.pow(y_pred, xp.asarray(1 - p, device=device_)) / (1 - p)
+            + xp.pow(y_pred, xp.asarray(2 - p, device=device_)) / (2 - p)
         )
     return float(_average(dev, weights=sample_weight))