manylinux1 fails on simple_on_word2vec
Created by: luotao1
[11:32:00] ++ for use_gpu in '$use_gpu_list'
[11:32:00] ++ ./simple_on_word2vec --dirname=/paddle/build/python/paddle/fluid/tests/book/word2vec.inference.model --use_gpu=true
[11:32:00] WARNING: Logging before InitGoogleLogging() is written to STDERR
[11:32:00] W0716 11:32:00.279105 27581 init.cc:82] Compiled with WITH_GPU, but no GPU found in runtime.
[11:32:00] W0716 11:32:00.294072 27581 init.cc:98] Compiled with WITH_GPU, but no GPU found in runtime.
[11:32:00] terminate called after throwing an instance of 'paddle::platform::EnforceNotMet'
[11:32:00] what(): Cannot open file /paddle/build/python/paddle/fluid/tests/book/word2vec.inference.model/__model__ at [/paddle/paddle/fluid/inference/io.cc:49]
[11:32:00] PaddlePaddle Call Stacks:
[11:32:00] 0 0xc91c85p paddle::platform::EnforceNotMet::EnforceNotMet(std::__exception_ptr::exception_ptr, char const*, int) + 583
[11:32:00] 1 0xd0d17bp paddle::inference::ReadBinaryFile(std::string const&, std::string*) + 811
[11:32:00] 2 0xd0f31ep paddle::inference::Load(paddle::framework::Executor*, paddle::framework::Scope*, std::string const&) + 222
[11:32:00] 3 0xc98187p paddle::NativePaddlePredictor::Init(std::shared_ptr<paddle::framework::Scope>) + 823
[11:32:00] 4 0xc98d92p std::unique_ptr<paddle::PaddlePredictor, std::default_delete<paddle::PaddlePredictor> > paddle::CreatePaddlePredictor<paddle::NativeConfig, (paddle::PaddleEngineKind)0>(paddle::NativeConfig const&) + 1762
[11:32:00] 5 0xc8e181p paddle::demo::Main(bool) + 225
[11:32:00] 6 0xc8f01ep main + 49
[11:32:00] 7 0x7feefdea1d1dp __libc_start_main + 253
[11:32:00] 8 0xc8df01pThe reason is that simple_on_word2vec demo depends on ctest -R test_word2vec -V unit-test. Thus, in manylinux1, we should not test simple_on_word2vec.