Created by: xiegegege

neural_machine_translation/rnn_search预测有时会报错(非必现)，报错如下： Traceback (most recent call last): File "infer.py", line 137, in infer() File "infer.py", line 113, in infer return_numpy=False) File "/usr/local/lib/python3.7/dist-packages/paddle/fluid/executor.py", line 525, in run use_program_cache=use_program_cache) File "/usr/local/lib/python3.7/dist-packages/paddle/fluid/executor.py", line 591, in _run exe.run(program.desc, scope, 0, True, True) paddle.fluid.core.EnforceNotMet: Invoke operator fill_constant error. Python Callstacks: File "/usr/local/lib/python3.7/dist-packages/paddle/fluid/framework.py", line 1317, in append_op attrs=kwargs.get("attrs", None)) File "/usr/local/lib/python3.7/dist-packages/paddle/fluid/layer_helper.py", line 56, in append_op return self.main_program.current_block().append_op(*args, **kwargs) File "/usr/local/lib/python3.7/dist-packages/paddle/fluid/layers/tensor.py", line 386, in fill_constant stop_gradient=True) File "/usr/local/lib/python3.7/dist-packages/paddle/fluid/contrib/decoder/beam_search_decoder.py", line 651, in early_stop shape=[1], value=0, dtype='bool', force_cpu=True, out=self._cond) File "/usr/local/lib/python3.7/dist-packages/paddle/fluid/contrib/decoder/beam_search_decoder.py", line 722, in decode self.early_stop() File "/ssd1/xiege/model_3.6/models-develop/fluid/PaddleNLP/neural_machine_translation/rnn_search/attention_model.py", line 215, in seq_to_seq_net decoder.decode() File "infer.py", line 60, in infer max_length=args.max_length) File "infer.py", line 137, in infer() C++ Callstacks: op fill_constant does not have kernel for data_type[bool]:data_layout[ANY_LAYOUT]:place[CUDAPlace(0)]:library_type[PLAIN] at [/paddle/paddle/fluid/framework/operator.cc:943]