enas
enas copied to clipboard
coordinator stopped with threads still running
Hi, I'm running the cifar10_macro_search.sh file and it seems that everything is fine because the training process goes well until the end, there in the last epcoh I get :
Epoch 310: Eval Eval at 109120 valid_accuracy: 0.7944 Eval at 109120 test_accuracy: 0.7890 I0320 07:33:09.999876 70367132797184 coordinator.py:393] Coordinator stopped with threads still running: QueueRunnerThread-dummy_queue-sync_token_q_EnqueueMany Exception in thread QueueRunnerThread-dummy_queue-sync_token_q_EnqueueMany (most likely raised during interpreter shutdown)
I'm using python 2.7.12
and tensorflow 1.12.0
. Any suggestions on how to fix this issue?
I think it might be because of tensorflow version as the code supports 1.4,1.7 . Can you try it with any of these versions?
Hi, I know is suppose to work just for 1.4, 1.7, but I was wondering how to make it work for a newer version of Tensorflow. And no, at the moment I cannot try with the older versions. Thanks, anyway.
Hi @lazarusA did you get a solution ? I have a same issue.
Hi, I know is suppose to work just for 1.4, 1.7, but I was wondering how to make it work for a newer version of Tensorflow. And no, at the moment I cannot try with the older versions. Thanks, anyway. did you get a solution ?
@rifaadev this is almost 2 years old :D. And at the end I didn't used it. But if I remember correctly I did got it to work installing all the right dependencies in the older version. No the new one.
@rifaadev this is almost 2 years old :D. And at the end I didn't used it. But if I remember correctly I did got it to work installing all the right dependencies in the older version. No the new one.
thank you for answer, I can't install an old version of tensorflow. do you know how?
I see the message in attached picture.