my brother in christ please take pity on my lack of backlinks. that said i do genuinely believe this, which is why i wrote it, and also want to hear the opinion of others (so i can make something useful).
1) Completion. There are still many tasks that can't be solved with a basic prompt.
2) Correctness. Subtle mistakes the AI might have missed, did you get it to handle edge cases? Did it break something else in the background?
3) Quality. This one is still a little handwavy, but there are a number of heuristics that you can use. These don't work across the board but you can design problems in a way that some solutions are clearly cleaner than others.
Maybe you could start by replying to my email? I would not have complained publicly if there had been any form of communication from you but there was none.
This is a reasonable criticism. I don't know how I can prove it as screenshots are not hard to fake, but here they are in any case. Also I emailed them and still haven't gotten a response, so I'm not inclined to be very understanding seeing as this appears to be happening quite frequently to people.
my company site is infoproc dot co. it is currently down. but if you google it you can see it has been indexed. And you can search for my company on linkedin if you want to know who I am.
Thanks for this tip. I tried it and I keep getting the error:
"Error: Your account got flagged by our internal fraud protection service" even after verifying my credit card.
This looks very interesting! I'm trying to understand if the flattening technique might work for my ts.
It's structured as follows:
At each time step t, I have an m by n data matrix. The value for m (rows) varies per time step. n stays constant and represents the features. And i want to predict one of the n values.
(In this case, t represents a single day, m (rows) represent the people that entered a store on that day, and n (cols) represent various features of the people. I want to predict one of those features, given the others.)
The fact that it's a time series matters, because i expect the relationship to change over time. For instance some feature n[x] (person wears a yellow shirt) might be correlated with my target feature n[y] (person steals) but only in the summer.
would it be possible to flatten this too? What would that look like?