Better models, bigger scale, easier to use
Show this thread
I'm pretty excited about the new workflows that become possible at the intersection of Keras preprocessing layers (end-to-end models), KerasTuner (hyperparameter tuning), and TF Cloud (large scale training in the cloud in just one line) ML tooling remains a fast-advancing area
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Keras Tuner is really amazing. Love. It.
-
Glad it's useful to you! And if you have any feedback on how to make it better, please share
- Show replies
New conversation -
-
-
This Tweet is unavailable.
-
I'm not really familiar with the course, but I believe one of the makers of the course is also the author of Trax. So it would make sense for them to use their own tool to teach their own course
End of conversation
-
-
-
Question about preprocessing and embedding layers: Can I move the weights from an embedding layer to a preprocessing layer? If I'm not going to train the embeddings further it would be nice to get rid of the memory allocation for gradients.
-
Sure, you can set layer.trainable = False on your embedding layer, or otherwise put the layer in your data pipeline instead of your model
End of conversation
New conversation -
-
-
Great to hear...just wanted to know if your book on Deep learning using python will be published soon...i loved the first edition..waiting for 2nd one
-
You could buy the MEAP from
@ManningBooks and you will get access to the chapters as he writes.https://www.manning.com/books/deep-learning-with-python-second-edition?query=Deep%20learning%20with%20python … - Show replies
New conversation -
-
-
Sorry if slightly off-topic, but how is TF 2.0 behaving now and/or how is its roadmap regarding official, upstream support for non-CUDA GPUs (AMD’s ROCm and foundational work for Intel’s oneAPI) in tensorflow-gpu?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.