tensorboard --logdir=runs not working: Abort trap: 6

问题

I am trying to run tensorboard: tensorboard --logdir=runs.
I have also tried: tensorboard --logdir=runs --host=127.0.0.1.
I am running the command from the terminal from within the the directory, which contains the runs folder.

I get the following error:

[libprotobuf FATAL external/com_google_protobuf/src/google/protobuf/descriptor.cc:1367] 
CHECK failed: GeneratedDatabase()->Add(encoded_file_descriptor, size): 
libc++abi.dylib: terminating with uncaught exception of type google::protobuf::FatalException: 
CHECK failed: GeneratedDatabase()->Add(encoded_file_descriptor, size): 
Abort trap: 6

My Python code contains the following lines:

tb_path = './runs/SimpleLSTM_MNIST'
if os.path.isdir(tb_path):
    shutil.rmtree(tb_path)

writer = tb.SummaryWriter(log_dir=tb_path)

My runs folder contains the folder SimpleLSTM_MNIST, which contains events.out.tfevents.1591953948.computername.local.29440.0.

Operating System: MacOS Catalina

How can I resolve the problem?

回答1:

looks like protobuf is not working properly, try install latest version

pip3 uninstall protobuf
pip3 install protobuf

回答2:

Apparently this is a specific issue that occurs when running macOS Catalina, and can be solved by switching to protobuf version 3.8.0 and tensorflow version 2.0.0.
So basically uninstalling tensorflow and protobuf and re-installing with pip3 install protobuf==3.8.0 and pip3 install tensorflow==2.0.0.

来源：https://stackoverflow.com/questions/62342221/tensorboard-logdir-runs-not-working-abort-trap-6

标签

python

pytorch

tensorboard

abort