我正在使用网络摄像头在Raspberry Pi 3 Model B +上设置TensorFlow对象检测API，以进行实时对象识别并获得以下运行时警告：
RuntimeWarning：'模块'tensorflow.python.framework.fast_tensor_util'的编译时版本3.4。如何在object_detection_picamera.py中解决此RuntimeWarning问题？
我已经按照this教程来检测实时对象。
pi@raspberrypi:~/tensorflow1/models/research/object_detection $ python3 Object_detection_picamera.py
/usr/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning:compiletime version 3.4 of module 'tensorflow.python.framework.fast_tensor_util' does not match runtime version 3.5
returnf('*args,**kwds')
/usr/lib/python3.5/importlib/_bootstrap.py:222: RuntimeWarning:builtins.type size changed, may indicate binary incompatibility. Expected 432, got 412
returnf('*args,**kwds')
2019-04-04 07:57:10.246546: F tensorflow/core/framework/op.cc:81] Non-OK-status:
ValidateKernelRegistrations(*this) status: invalid argument: HostMemory arg 'r1' not found in OpDef:Op<name=BroadcastGradientArgs; signature=s0:T, s1:T -> r0:T, r1:T; attr=T:type, default=DT_INT32,allowed=[DT_INT32, DT_INT64]>
Aborted
该代码在几个月前运行良好，但现在不断收到此运行时警告。