3

I am encountering a problem where my Kernal automatically restarts when i run ML model on Spyder IDE I have checked other questions concerning the same topic and none of the answers could solve my situation, any solution for this problem would be greatly appreciated.

Following is the output from ipython console

  Allocation of 37748736 exceeds 10% of free system memory.
    2021-06-26 23:41:03.711755: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 37748736 exceeds 10% of free system memory.
WARNING:tensorflow:Can save best model only with val_accuracy available, skipping.

Restarting kernel...
DevanDev
  • 161
  • 1
  • 2
  • 17
  • I also got same error, does anyone have idea ? It stop execution in between and giving this error i console, Restarting kernel... – Kiran Nov 01 '21 at 07:20
  • Hi! Could you check out these similar threads for answer ? https://stackoverflow.com/questions/61505749/tensorflowcan-save-best-model-only-with-val-acc-available-skipping https://stackoverflow.com/questions/50304156/tensorflow-allocation-memory-allocation-of-38535168-exceeds-10-of-system-memor –  Nov 11 '21 at 02:20
  • @TensorflowSupport , Thanks for your response, i tried it but still i get similar error – DevanDev Nov 11 '21 at 07:23
  • Could you share a stand alone code to replicate the issue then? –  Nov 11 '21 at 08:39
  • @TensorflowSupport , i am training my model for 1500 epochs , i have more than 35000 images and the code is here https://stackoverflow.com/questions/68162820/how-to-perform-super-pixel-image-segmentation-and-feature-extraction – DevanDev Nov 11 '21 at 09:00
  • Ok @DevanDev! Could you check the suggestion in above thread aside below threads and see if the issue persits! https://github.com/tensorflow/tensorflow/issues/33163 https://github.com/tensorflow/tensorflow/issues/44107 –  Nov 15 '21 at 12:28

0 Answers0