hidden bcast_params call in dist train (#11575)

Author: Yancey

benchmark/fluid/fluid_benchmark.py

@@ -264,8 +264,6 @@ def train_parallel(avg_loss, infer_prog, optimizer, train_reader, test_reader,
                     break
             else:
                 loss, = exe.run([avg_loss.name], feed=feeder.feed(data))
-            if args.update_method == "pserver":
-                exe.bcast_params()
             if args.use_reader_op:
                 num_samples += args.batch_size * args.gpus
             else:

python/paddle/fluid/parallel_executor.py

@@ -71,7 +71,6 @@ def __init__(self,
                  num_trainers=1,
                  trainer_id=0,
                  **kwargs):
-
         if len(kwargs) != 0:
             err_msg = ""
             for key in kwargs:
@@ -130,6 +129,11 @@ def __init__(self,
         main = main_program
         main = main if main else framework.default_main_program()
         scope = executor.global_scope()
+        # FIXME(Yancey1989): it's a temporary approach to determinate the distribute
+        # train program, call self.bcast_param() at the end of each mini-batch.
+        self.is_dist = True if "recv" in [
+            op.type for op in main.global_block().ops
+        ] else False
 
         if share_vars_from and not isinstance(share_vars_from,
                                               ParallelExecutor):
@@ -262,6 +266,10 @@ def run(self, fetch_list, feed=None, feed_dict=None):
         fetch_var_name = '@FETCHED_VAR_NAME@'
         self.executor.run(fetch_list, fetch_var_name)
         arr = self.scope.find_var(fetch_var_name).get_lod_tensor_array()
+
+        if self.is_dist:
+            self.bcast_params()
+
         return [arr[i] for i in range(len(arr))]
 
     def bcast_params(self):