71–73 of 73 posts
As someone who has dealt with countless breaking changes in keras and wasted days of my life attempting to upgrade, no thank you. My pytorch code from years ago still works with no issues, my old keras code would break all the time even in minor releases.
Earlier quoted context omitted.
The large Keras userbase exists largely because Tensorflow sucked.
I'll agree to disagree. I find great value in the Keras API. It's also a bit histrionic, in that Keras was very popular with the Theano backend before that project wound down.
When I switched to TF in 2016 Keras was still in its infancy so I wrote a lot of low level TF code (eg my own batchnorm layer), but many new DL researchers struggled with TF because they didn’t have Theano experience. That steep learning curve led to the rise of Keras as the friendly TF interface.
Pytorch is a different story.
Supporting multiple backends (especially Jax) is nice! Makes experimenting/migrating between them so much more approachable. Any timeline on when can we expect support for distributed Jax training? The doc currently seems to indicate only TF is supported for distributed training.
Support for distributed JAX training demoed here: bit.ly/keras-on-jax-demo You have to write a custom training loop for now, but it works.