migrate existing log callsites

Summary:
**Diff Purpose & Changes**
Now that we have the library log function, let's migrate existing usages over.

Reviewed By: colin2328

Differential Revision: D72941480

fbshipit-source-id: a39e1dfad8e9f6dfd0d2c8ee429b04ed04065bb1

Author: amirafzali

python/monarch/parallel/pipelining/runtime.py

@@ -35,8 +35,6 @@
 logger = getLogger()
 
 
-log = remote("monarch.worker.worker.log", propagate="inspect")
-
 run_forward_udf = remote(
     "monarch.parallel.pipelining.runtime.run_forward_impl",
     propagate=lambda stage, input_tensor, model_chunk_id, microbatch_id: input_tensor,

python/monarch/worker/worker.py

@@ -54,10 +54,6 @@
     CONTROLLER_COMPILED_REPEAT = True
 
 
-def log(*args):
-    logger.warning(*args)
-
-
 def set_worker_logging_level(level):
     logger.setLevel(level)
 

python/tests/test_coalescing.py

@@ -35,8 +35,6 @@ def _do_bogus_tensor_work(x, y, fail_rank=None):
     propagate=_do_bogus_tensor_work,
 )
 
-log = remote("monarch.worker.worker.log", propagate="inspect")
-
 
 def inspect(x):
     return fetch_shard(x).result().item()

python/tests/test_remote_functions.py

@@ -23,6 +23,7 @@
     Stream,
 )
 from monarch._testing import BackendType, TestingContext
+from monarch.builtins.log import log_remote
 from monarch.cached_remote_function import remote_autograd_function
 from monarch.common import remote as remote_module
 from monarch.common.device_mesh import DeviceMesh
@@ -60,9 +61,6 @@ def custom_excepthook(exc_type, exc_value, exc_traceback):
 sys.excepthook = custom_excepthook
 
 
-log = remote("monarch.worker.worker.log", propagate="inspect")
-
-
 def _set_device_udf(*args):
     return torch.zeros(1)
 
@@ -306,7 +304,7 @@ def do_test_all_reduce(cls, backend_type, reduce_op, expected_tensor):
 
     def test_hello(self, backend_type):
         with self.local_device_mesh(2, 2, backend_type):
-            log("hello, world")
+            log_remote("hello, world")
 
     def test_eager_remote_function_failed(self, backend_type):
         if backend_type == BackendType.PY:
@@ -335,15 +333,15 @@ def test_simple_tensors(self, backend_type):
         with self.local_device_mesh(2, 2, backend_type):
             x = torch.rand(3, 4)
             y = x + x
-            log("%s %s", x, y)
+            log_remote("%s %s", x, y)
             z = torch.std_mean(x)
-            log("%s", z)
+            log_remote("%s", z)
 
     def test_user_call(self, backend_type):
         with self.local_device_mesh(2, 2, backend_type) as _:
             x = torch.rand(3, 4)
             y = rlist((x + 1, x))
-            log("%s", y)
+            log_remote("%s", y)
 
             # resume monday:
             # 1. tensor ctor resource guard (done)