This is insanely cool
This very simply Kaggle notebook shows how to handle structured data with Keras preprocessing layers and how to select a high-performing model using KerasTuner:https://www.kaggle.com/fchollet/titanic-keras-kerastuner-best-practices …
-
-
-
How to download the notebook and data set files?
- Show replies
New conversation -
-
-
Soo cool....
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thank you for sharing
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Hi Francois, thx for sharing. I am wondering if the last Dense layer miss the activation='sigmoid'? The outputs are not just 0 and 1 right now :(
-
There probably should be a 'sigmoid' there. I think binary cross entropy loss clips to (0,1) and accuracy uses True>0.5 so optimization and loss works, but submission is wrong.
- Show replies
New conversation -
-
-
@aureliengeron@__mharrison__@mlpowered With this notebook, can I run run ? joblib.dump(train_ds, './titanic_ds.joblib') -
It is giving following error: InternalError: Tensorflow type 21 not convertible to numpy dtype. How to fix this error please? Thanks.
- Show replies
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.