From the first example:
> return -tf.reduce_mean((T("input") - 0.5)*2)
Don't you want to use the average instead of "0.5"?
From the explanation:
> The code is a function that takes in an input tensor and returns the mean squared error of the input.
It's hard to believe that the system can detect that it's the "mean squared error". Is this an explanation a real example from the implementation, or it's just a handmade example?
The live example on the home page doesn't seem to work. I assumed JS was supported and typed this:
const addOne = function(someNumber) { return someNumber + 1 }
The "explain" box shows nothing.
This site makes bold claims which, so far, I don't see backed up by anything. Even the "join waitlist" dialog is broken!
Next...