AttributeError when running a translator

[TonyWild]

Hi,
I'm trying to run the demo example, and run the command
python translator/java-dnn-gen.py --model kerasmodels/mnist-lowquality.h5 --outputs "<path to jpf-symbc>/jpf-symbc/src/examples/neurospf" -d demoimage as in readme.md

However, when I trying to run the command, this error happens:
AttributeError: 'int' object has no attribute 'value'
in line 47: prog+=' int run(double[] input)\n'.format(input_shape[1].value)

So when I removed .value from the line 47, the error does not appear at line 47, but the new errors appear at different line, related to this. (_out.shape[1].value also causes a same AttributeError)

Can you check this problem? I'm using tensorflow of version 2.5.0-rc1.
Thank you!

[muhammadusman93]

Hi Tony, Can you try with Keras version 2.2.4 and Tensorflow version 1.13.1? M. Usman

…

On Thu, Apr 15, 2021 at 5:20 AM TonyWild ***@***.***> wrote: Hi, I'm trying to run the demo example, and run the command python translator/java-dnn-gen.py --model kerasmodels/mnist-lowquality.h5 --outputs "<path to jpf-symbc>/jpf-symbc/src/examples/neurospf" -d demoimage as in readme.md However, when I trying to run the command, this error happens: AttributeError: 'int' object has no attribute 'value' in line 47: prog+=' int run(double[] input)\n'.format(input_shape[1].value) So when I removed .value from the line 47, the error does not appear at line 47, but the new errors appear at different line, related to this. ( _out.shape[1].value also causes a same AttributeError) Can you check this problem? I'm using tensorflow of version 2.5.0-rc1. Thank you! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHQQSAANOSLMGVJT7UWWIV3TI24WRANCNFSM427FX2FA> .