I am trying to load a keras model to do classification task using my raspberry. But after running this code:

import keras
from keras.models import load_model

model = load_model("mobilenet_v2_3(trained).h5")

before i will be able to do some prediction, i got this error;

Using TensorFlow backend.
Instructions for updating:
Colocations handled automatically by placer.
Backend terminated(returncode: -11)
Fatal python error: Segmentation fault 
  • Bad luck. So you have a segFault. Most likely your big data is too big. Ref: Identify what's causing segmentation faults kb.iu.edu/d/aqsj
    – tlfong01
    Commented May 23, 2019 at 1:26
  • Do you suggest a kind of solution Please @tlfong01
    – Go Deeper
    Commented May 23, 2019 at 7:17
  • Well, I am a small guy, don't know nothing about big data, not even scratching surface knowledge. But now everybody is talking about AI, Neural Net and TensorFlow, etc. I am losing face because I can't even tell what is the difference between Neural Net and TensorFlow. So I am keen to dig deeper below the surface, memorize more weird names, so that my friends would respect more than I deserve. My plan is first wiki, and pretend to give an answer. Perhaps you can also pretend to give an answer, then I upvote you, and upvote me, building our reputation together, ... :)
    – tlfong01
    Commented May 23, 2019 at 8:00


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.