feat: add cpu function

functional/frontend/torch/torch.py

@@ -0,0 +1,19 @@
+# in ivy/functional/frontends/torch/torch.py
+@to_ivy_arrays_and_back
+
+import torch
+import ivy.functional.frontends.torch as torch_frontend
+
+def to_cpu(self):
+     if self.is_cuda:
+          return torch_frontend.self.to_cpu()
+     else:
+          return self
+
+torch.Tensor.to_cpu = to_cpu
+
+x = torch.randn(2, 3).cuda()
+
+y = x.to_cpu()
+
+print(y.device)

ivy_tests/test_ivy/test_frontends/test_torch/test_tensor

@@ -0,0 +1,45 @@
+from ivy_test.helpers import *
+from ivy_test.helpers import helpers
+from ivy_test.helpers.helpers import handle_frontend_method
+
+@handle_frontend_method(
+  class_tree=CLASS_TREE,
+  init_tree="torch.tensor",
+  method_name="to_cpu",
+  dtype_and_x=helpers.dtype_and_values(
+     available_dtypes=helpers.get_dtypes("float"),
+       num_arrays=1,
+       min_value=-1e04,
+       max_value=1e04,
+       allow_inf=False,
+       ),
+)
+
+
+def test_torch_instance_to_cpu(
+  dtype_abd_x,
+  frontend,
+  backend_fw,
+  frontend_method_data,
+  init_flags,
+  method_flags,
+):
+  input_dtype, x = dtype_and_x
+  helper.test_frontend_method(
+    init_input_dtypes=input_dtype,
+    init_all_as_kwargs_np={
+      "data"= x[0],
+    },
+    method_input_dtypes=input_dtype,
+    method_all_as_kwarg_np={},
+    frontend_method_data=frontend_method_data,
+    init_flags=init_flags,
+    method_flags=method_flags,
+    frontend=frontend,
+    backed_to_test=backend_fw,
+  )
+
+
+
+# run the test
+test_torch_instance_to_cpu