Created by: xiegegege

在paddle1.5分支下，ctr训练有如下报错： Traceback (most recent call last): File "train.py", line 262, in train() File "train.py", line 232, in train train_loop(args, main_program, py_reader, loss, auc_var, batch_auc_var, 1, 0) File "train.py", line 155, in train_loop exec_strategy=exec_strategy) File "/opt/_internal/cpython-3.6.0/lib/python3.6/site-packages/paddle/fluid/parallel_executor.py", line 175, in init self._compiled_program._compile(place=self._place, scope=self._scope) File "/opt/_internal/cpython-3.6.0/lib/python3.6/site-packages/paddle/fluid/compiler.py", line 366, in _compile scope=self._scope) File "/opt/_internal/cpython-3.6.0/lib/python3.6/site-packages/paddle/fluid/compiler.py", line 337, in _compile_data_parallel self._exec_strategy, self._build_strategy, self._graph) paddle.fluid.core_avx.EnforceNotMet: If you set build_strategy.reduce with 'Reduce',the number of places must be greater than 1. at [/ssd1/xiege/paddle_ce/Paddle/paddle/fluid/framework/parallel_executor.cc:325]