search

nicknochnack / TFODCourse

963 stars 924 forks source link

AttributeError: module 'keras._tf_keras.keras.layers' has no attribute 'experimental' #174

Open Shanyageo opened 2 months ago

kayleedirecto commented 2 months ago

Hi there! Did you end up finding a solution to this?

davemaster commented 2 months ago

Hi there! Did you end up finding a solution to this?

You have to downgrade TensorFlow to 2.10 if I remember

kayleedirecto commented 2 months ago

Do you remember what version of Python you were running?

sabbir2609 commented 2 months ago

3.10.6 or 3.9.6 should work

davemaster commented 2 months ago

Do you remember what version of Python you were running?

Python 3.9.13

and here my packets installed myrequirements.txt

davemaster commented 2 months ago

SINCERELY, I'm questioning this python way to do Machine Learning, EVERY TIME, those packets needed to do so, every time, there are a lot of problems of compatibility, a MESS DISASTER what really, REALLY WASTE MORE TIME than that used to train a model... MUST BE a better way to sync these hundred of packets, or at least for 6 months, NO CHANGES at all.

kayleedirecto commented 2 months ago

I know, it's really frustrating :/ Were you able to get all the packages installed with that version combination?

sabbir2609 commented 2 months ago

SINCERELY, I'm questioning this python way to do Machine Learning, EVERY TIME, those packets needed to do so, every time, there are a lot of problems of compatibility, a MESS DISASTER what really, REALLY WASTE MORE TIME than that used to train a model... MUST BE a better way to sync these hundred of packets, or at least for 6 months, NO CHANGES at all.

Use Anaconda

davemaster commented 2 months ago

Anaconda, another stuff; then what, tiger, lion, rhino, spider.....

better do this using c++ (rules, forever) or c#, and no more wasting time with volatile packages ...

Good luck, AND THANKS anyway...