Place control of the pThread into the Summary process

mindspore/train/summary/_writer_pool.py

@@ -31,16 +31,6 @@ except ValueError:
     import multiprocessing as ctx
 
 
-# Environment variables are used to specify a maximum number of OpenBLAS threads:
-# In ubuntu(GPU) environment, numpy will use too many threads for computing,
-# it may affect the start of the summary process.
-# Notice: At present, the performance of setting the thread to 2 has been tested to be more suitable.
-# If it is to be adjusted, it is recommended to test according to the scenario first
-os.environ['OPENBLAS_NUM_THREADS'] = '2'
-os.environ['GOTO_NUM_THREADS'] = '2'
-os.environ['OMP_NUM_THREADS'] = '2'
-
-
 def _pack_data(datadict, wall_time):
     """Pack data according to which plugin."""
     result, summaries, step = [], [], None
@@ -78,6 +68,15 @@ class WriterPool(ctx.Process):
         self.start()
 
     def run(self):
+        # Environment variables are used to specify a maximum number of OpenBLAS threads:
+        # In ubuntu(GPU) environment, numpy will use too many threads for computing,
+        # it may affect the start of the summary process.
+        # Notice: At present, the performance of setting the thread to 2 has been tested to be more suitable.
+        # If it is to be adjusted, it is recommended to test according to the scenario first
+        os.environ['OPENBLAS_NUM_THREADS'] = '2'
+        os.environ['GOTO_NUM_THREADS'] = '2'
+        os.environ['OMP_NUM_THREADS'] = '2'
+
         with ctx.Pool(min(ctx.cpu_count(), 32)) as pool:
             deq = deque()
             while True: