2/ Like, I get the value of concrete examples. But is petal lengths really it? Besides, earning intuition for what you can and what is hard to recover is more difficult when the cool example was selected specifically because a method works well on it.
-
-
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Love this idea. I read some excellent books and articles on deep learning, but the walk-throughs were not directly applicable to my specific data and challenges, so I had to hammer through a lot of trial and error to figure out a path forward.
-
Also, I have ADHD, and I had some learning challenges with this stuff -- point being, love your idea and I had kind of similar thoughts as I was slogging through. It'd be cool to have "machine learning for ADHD nerds" or something. Ha!
- 1 more reply
New conversation -
-
-
Yeah I to create my own to learn a few different things specific to disease data. sometimes the routine examples just dont apply and you need it to be domain specific
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
After teaching bioinformatics— generating one’s own data in the process of learning analysis really helps- students better understand how to assess data types and can begin to see why different analyses suite different data sets. It also creates a sense of ownership as well.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Something that I’ve wanted to do is estimate a distribution via MLE, use the estimated parameters to generate training data, fit a model to that data, and then test the model. Not sure if that’s a thing, but it seems like that’s being done with backtesting investment strategies.
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.