I want to know if we can use GPT models to do task that normal Machine Learning or Deep Learning models do. For example, is it possible for us to predict employee churn based on previously churned employee information? Can GPT do these level of predictions?
Hi and welcome to the developer forum!
It’s a very interesting question, I’m not sure if one of the GPT models is the correct tool for this, it really depends on the magnitude of the data involved, in the past, “expert systems” had a limited capability to spot trending in data, and to an extent the GPT models can perform very well spotting irregularities and differences in patterns, but they are limited to a few 10’s of thousands of tokens.
As you mentioned, this sort of falls into the wheelhouse of a neural net with some loss function able to score successful detection of such.
The honest answer is maybe? give it a shot, I’d be super interested to see what you find out and how you go about formatting the data.
1 Like