chore: resolve lint

XuehaiPan · XuehaiPan · commit 3f466664d997 · 2022-11-03T00:14:39.000+08:00
diff --git a/examples/iMAML/imaml_omniglot.py b/examples/iMAML/imaml_omniglot.py
@@ -51,7 +51,7 @@ class InnerNet(
     def __init__(self, meta_net, n_inner_iter, reg_param):
         super().__init__()
         self.meta_net = meta_net
-        self.net = copy.deepcopy(meta_net)
+        self.net = torchopt.module_clone(meta_net, by='deepcopy', detach_buffers=True)
         self.n_inner_iter = n_inner_iter
         self.reg_param = reg_param
 
diff --git a/tests/test_implicit.py b/tests/test_implicit.py
@@ -243,7 +243,7 @@ class InnerNet(ImplicitMetaGradientModule, has_aux=True):
         def __init__(self, meta_model):
             super().__init__()
             self.meta_model = meta_model
-            self.model = copy.deepcopy(meta_model)
+            self.model = torchopt.module_clone(meta_model, by='deepcopy', detach_buffers=True)
 
         def forward(self, x):
             return self.model(x)
diff --git a/torchopt/diff/implicit/nn/module.py b/torchopt/diff/implicit/nn/module.py
@@ -81,9 +81,9 @@ def enable_implicit_gradients(
         raise ValueError('Implicit gradients are already enabled for the solve function.')
 
     cls_has_aux = cls.has_aux
-    custom_root_kwargs = dict(has_aux=cls_has_aux)
-    if cls.linear_solve is not None:
-        custom_root_kwargs.update(solve=cls.linear_solve)
+    custom_root_kwargs = dict(has_aux=cls_has_aux, solve=cls.linear_solve)
+    if cls.linear_solve is None:
+        custom_root_kwargs.pop('solve')
 
     @functools.wraps(cls_solve)
     def wrapped(  # pylint: disable=too-many-locals
@@ -145,7 +145,7 @@ def optimality_fn(
                 ):
                     container.update(container_backup)
 
-        @custom_root(optimality_fn, argnums=1, **custom_root_kwargs)
+        @custom_root(optimality_fn, argnums=1, **custom_root_kwargs)  # type: ignore[arg-type]
         def solver_fn(
             flat_params: TupleOfTensors,  # pylint: disable=unused-argument
             flat_meta_params: TupleOfTensors,  # pylint: disable=unused-argument
