Inference is the term used to represent the consumption of compute by AI model to answer. Models are trained at something but inference is when they use their compute to actually do it.
You can compare it with you and your capacity to drive. You know what the car will do at any of your actions. But when you actually drive, you take input (your vision) think about it and output decisions (your actions to move the car depending on what is actually happening).
This is inference for an AI model, and it is exploding as more & more trained model are deployed, to which we ask more & more tasks - from answering questions to drive autonomously, passing by generating images or sounds.
Can you explain what inference means in this context..? Google translate doesn't make sense
Inference is the term used to represent the consumption of compute by AI model to answer. Models are trained at something but inference is when they use their compute to actually do it.
You can compare it with you and your capacity to drive. You know what the car will do at any of your actions. But when you actually drive, you take input (your vision) think about it and output decisions (your actions to move the car depending on what is actually happening).
This is inference for an AI model, and it is exploding as more & more trained model are deployed, to which we ask more & more tasks - from answering questions to drive autonomously, passing by generating images or sounds.
Better?
Thanks for the review. Need to be patient for a bargained stock price to arrive.
Always. Patience is the hardest part of the job