What's a ML problem you are interested in but for which you haven't found a good tutorial / introduction?
-
-
Also, if you are working with a specific dataset, please describe it
Show this thread -
Lots of excellent ideas in this thread! Thanks to everyone who's contributing :)
Show this thread
End of conversation
New conversation -
-
-
Online learning
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Would be interesting to see in a tutorial different ways to cheat a neuronal net and how to prevent that. Havent done research if smt like that is outhttps://towardsdatascience.com/breaking-neural-networks-with-adversarial-attacks-f4290a9a45aa …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Take a look at ML security
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I have a huge database of objects with string names (and - sometimes - other metadata). During inference, an unknown object comes in, denoted by its string name (in slight variations) and - sometimes - bit of metadata. /1
-
Now I want to match the incoming object to the correct object in my database without comparing it with every single object in the database. I also want to know when such a match does not exist. How do I ideally structure this problem setup? I can use non-ML tools. /2
End of conversation
New conversation -
-
-
am training a machine to learn to draw art from line sketches from the scratch.. Am using Google datasets called quick draw
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
to train the model from scratch but the output has not improved.. What model can u i use apart from the RNNs and LSTM by Google?
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.
