Getting an error

when i am trying to run my project i am getting very cheeky error. I have expored across a lot of solutions , but i am unable to resolve the error, plz if you can check for the error. I have already spent about a day over the issue,
error that i am encountering is :-

AttributeError: module ‘tensorflow.python.pywrap_tensorflow’ has no attribute ‘PyExceptionRegistry_Init’

so please check for the error in my code. Please clear my issue over live chat …it would be of great help , or if you can share you contact over the live chat…Hope to see your response soon!!

hey @saksham_thukral ,
try installing latest version of tensorflow again. My be it would workout.

tried that also , it didn’t worked! please if you can resolve my issue over live chat, it would be of great help?

hey @saksham_thukral , there is no such need of live chat as there is some problem in system files of tensorflow and you need to try a lot for it to work.

which tensorflow version are you currently working on ? and have you downloaded tensorflow-gpu or tensorflow?

my tensorflow version is 1.14.0 and how can i check whether tensorflow-gpu is running or not?Sir actually i am stuck here from quite a long time now, as my course duration is passing by,so please if you can check my issue over team-viewer,it would be of great help. :worried:

hey @saksham_thukral ,
see this is not a task that can be done on team viewer easily , it can take time to be resolved. till then i would recommend you to use google colab or kaggle kernels for practice . I am trying to find a solution for this. These are some system problems and require many hit and tries to solve it.

try installing tensorflow 1.15.0
using command pip install tensorflow==1.15

i tried uninstalling the whole tensorflow and then re-installed the tensorflow 2.2.0 version with gpu support, so now my server although works but know when i give the sentence, and click on predict button i get the following error?


Can u please address this issue?

hey @saksham_thukral ,
go for tensorflow version 1.15 , tensorflow 2.2.0 currently is a bit unstable and has many changes , it will a whole lot of work to work according to it.

just once try with 1.15 and let me if it works.

what i did was according to the tensorflow git issue as follows:

, and that although removed my this error, but know i even tried to to run only api.py, even it ran correctly without error, but didn’t showed the emoji in output in terminal as in video at 9:33,

from one doubt to another , its really good.
Try printing something at the code line where you are implementing to show the emoji.
A kind of debugging , it will help you to find where’s the error.

Even if you can’t find out, then kindly share your code and other files with me.

now i tried after reinstalling version 1.15 of both tensorflow and tensorflow gpu, but it gave some differrent error,please if you check my code? And how do you want me to share my code with you (via github?), or isn’t it better if you just check my code over team viewer? But you can tell me either way you are comfortable. But please i really want to resolve this issue ASAP, as i am stuck here now from more than 4 days now, and my course is just passing by, …

Finally the error that i am getting on running my server after using 1.15 version is as follows:

hey @saksham_thukral ,
Currently i dont have such good internet connection for team viewer , you can whatsapp me at 8630831390 or can even call me at the same if there is something urgent you need.

yes

yeah it is, but problem with internet connection , for me github will be better.