I think the JAX team has been doing a great job, and the growing popularity of JAX in the deep learning research community reflects it!
Use http://keras.io . Buy my book (you can buy the preview version of the 2nd edition already), look at the developer guides at https://keras.io/getting_started/ … and at the code examples at https://keras.io/examples/ . Most effective way to get started.
-
-
As we become more advanced what should be the preference? I have been working with keras from the past 2 years! Now I want a lot more customisation! Any good resources for pure tf2?
-
Check out our developer guides, such as https://keras.io/guides/making_new_layers_and_models_via_subclassing/ … orhttps://keras.io/guides/customizing_what_happens_in_fit/ …
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.