Model visualization for covariance between input variables


I have built a VAE model dealing with tabular oceanographical data. My question is, what should I do to
visualize model weight and clarify the covariance relationship between input variables, as shown in the following chart.

Here is the website of the screenshot.

Please note that I don’t need it to be dynamic; I only need to visualize the final model.


Hi @hawkiyc! Thank you very much for writing in!

To confirm you are interested in visualizing model weight and clarifying the covariance relationship between input variables in Tensorflow, not in W&B?


Hi @artsiom,

My data are oceanographical records, and I am trying to clarify the covariance relationship between all inputs. I have already built a VAE with PyTorch, and now I am wondering if it is possible to visualize the weight as the chart I have posted.



Hi @hawkiyc! Happy New Year and apologies for the wait!

Looking into this topic it a bit and unfortunately, this falls out of our scope and we aren’t able to help you much with PyTorch concerns.

If you do ever have any questions regarding our platform we’ll be happy to help!
I will close this request internally from wandb’s side but will keep it open here on the forums, for community members to provide feedback