我正在尝试在程序中加载keras（Tensorflow后端）模型。但是，导入keras时似乎立即失败：
from keras.models import load_model
下面是lldb中的调试跟踪：
Process 26442 stopped
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=EXC_I386_GPFLT)
frame #0: 0x0000000113915eee libtensorflow_cc.so`tensorflow::OpDef::SharedDtor() + 158
libtensorflow_cc.so`tensorflow::OpDef::SharedDtor:
-> 0x113915eee <+158>: jmpq *0x8(%rax)
0x113915ef1 <+161>: popq %rbx
0x113915ef2 <+162>: popq %r14
0x113915ef4 <+164>: popq %rbp
Target 0: (Python) stopped.
(lldb) bt
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=EXC_I386_GPFLT)
* frame #0: 0x0000000113915eee libtensorflow_cc.so`tensorflow::OpDef::SharedDtor() + 158
frame #1: 0x0000000113915cdc libtensorflow_cc.so`tensorflow::OpDef::~OpDef() + 28
frame #2: 0x000000012d1ba71c _pywrap_tensorflow_internal.so`_GLOBAL__sub_I_test_construction_fails_op.cc + 76
frame #3: 0x0000000108125cc8 dyld`ImageLoaderMachO::doModInitFunctions(ImageLoader::LinkContext const&) + 518
frame #4: 0x0000000108125ec6 dyld`ImageLoaderMachO::doInitialization(ImageLoader::LinkContext const&) + 40
frame #5: 0x00000001081210da dyld`ImageLoader::recursiveInitialization(ImageLoader::LinkContext const&, unsigned int, char const*, ImageLoader::InitializerTimingList&, ImageLoader::UninitedUpwards&) + 358
frame #6: 0x0000000108120254 dyld`ImageLoader::processInitializers(ImageLoader::LinkContext const&, unsigned int, ImageLoader::InitializerTimingList&, ImageLoader::UninitedUpwards&) + 134
frame #7: 0x00000001081202e8 dyld`ImageLoader::runInitializers(ImageLoader::LinkContext const&, ImageLoader::InitializerTimingList&) + 74
frame #8: 0x0000000108112d20 dyld`dyld::runInitializers(ImageLoader*) + 82
frame #9: 0x000000010811c75f dyld`dlopen_internal + 609
frame #10: 0x00007fff75ef05f3 libdyld.dylib`dlopen + 200
fra
me #11: 0x00000001057dba0e Python`_PyImport_FindSharedFuncptr + 302
frame #12: 0x00000001057b0d36 Python`_PyImport_LoadDynamicModuleWithSpec + 598
frame #13: 0x00000001057b0799 Python`_imp_create_dynamic + 329
frame #14: 0x00000001056d2038 Python`_PyMethodDef_RawFastCallDict + 248
frame #15: 0x00000001056d1d21 Python`PyCFunction_Call + 209
frame #16: 0x000000010578da08 Python`_PyEval_EvalFrameDefault + 26360
frame #17: 0x00000001057910d6 Python`_PyEval_EvalCodeWithName + 2422
我有以下的keras和tensorflow版本：
Keras==2.2.4
Keras-Applications==1.0.8
Keras-Preprocessing==1.1.0
tensorboard==1.14.0
tensorboardX==1.8
tensorflow==1.14.0
tensorflow-estimator==1.14.0
知道我错过了什么吗？谢谢！