Parts of the Code does not run. Phyton said that apparently versions of tensorflow and keras do not match for this example.
@mubsi, maybe this needs a check-up?
Hi @Jaime4,
I ran the lesson on the learning platform, all the executions were successful and as expected.
Please reply with the following:
- share a screenshot of the code (and error) you are referring to
- where are you running the code ? DLAI’s learning platform ? or locally?
Thanks,
Mubsi
Hello,
I am running the code locally, using Jupyter Notebook.
The part of code that generates the error is:
kl_loss = 0.5 * K.sum(K.square(mu_flat) + K.exp(log_var_flat) - log_var_flat - 1, axis = -1)
vae_flat_loss = reconstruction_loss + kl_loss
Build model
Ensure that the reconstructed outputs are as close to the inputs
vae_flat = Model(inputs_flat, outputs_flat)
vae_flat.add_loss(vae_flat_loss)
vae_flat.compile(optimizer=‘adam’)
Also, the message error in included.
When you do that, you have to be extremely careful that your local tools and packages are exactly the same versions that the DLAI platform uses.
How can I know the tools versions that DLAI uses?
It seems that tensorflow required is 2.14.0, but this is not available now.
Perhaps you will have to modify the assignments so they work with the tools you have installed.
Hi @Jaime4,
There are the packages used for this course:
weaviate-client==3.24.2
openai==0.28.1
requests==2.31.0
numpy==1.26.0
networkx==3.1
matplotlib==3.8.0
tensorflow==2.14.0
scipy==1.11.3
sentence-transformers==2.2.2
seaborn==0.13.0
python-dotenv==1.0.0
contourpy==1.0.7
cycler==0.11.0
fonttools==4.39.3
packaging==23.1
pyparsing==3.0.9
python-dateutil==2.8.2
six==1.16.0
click
seaborn==0.13.0
kiwisolver==1.4.4
Pillow==9.5.0
Many, many thanks!!