Created by: langwenjing

使用python api训练模型，指定了paddle.init(use_gpu=True, trainer_count=4)，并且设置了CUDA_VISIBLE_DEVICES，启动发现gpu占用为0，请问还需要配置什么参数吗？ 并且程序运行一段时间出现中断：

* Aborted at 1512634078 (unix time) try "date -d @1512634078" if you are using GNU date *

PC: @ 0x0 (unknown)

* SIGFPE (@0x7f19e84f5e39) received by PID 14737 (TID 0x7f1a38005700) from PID 18446744073312099897; stack trace: *

@ 0x7f1a37bdc160 (unknown) @ 0x7f19e84f5e39 paddle::AucEvaluator::evalImp() @ 0x7f19e84f6805 paddle::Evaluator::eval() @ 0x7f19e84c2048 paddle::CombinedEvaluator::eval() @ 0x7f19e84c73f2 paddle::MultiGradientMachine::eval() @ 0x7f19e8377daa _wrap_GradientMachine_eval @ 0x4b4cb9 PyEval_EvalFrameEx @ 0x4b6b28 PyEval_EvalCodeEx @ 0x4b5d10 PyEval_EvalFrameEx @ 0x4b6b28 PyEval_EvalCodeEx @ 0x4b5d10 PyEval_EvalFrameEx @ 0x4b6b28 PyEval_EvalCodeEx @ 0x4b5d10 PyEval_EvalFrameEx @ 0x4b5fb8 PyEval_EvalFrameEx @ 0x4b6b28 PyEval_EvalCodeEx @ 0x4b6c52 PyEval_EvalCode @ 0x4e1c7d PyRun_FileExFlags @ 0x4e3501 PyRun_SimpleFileExFlags @ 0x4159dd Py_Main @ 0x7f1a37136bd5 __libc_start_main @ 0x414b71 (unknown) @ 0x0 (unknown)

请帮忙给出gpu训练方法，谢谢~