python - Where does machine learning algorithme store the result? -

i think kind of "blasphemy" comes ai world, since come world program , result, , there concept of storing un memory, here question :

machine learning works iterations, more there iterations, best our algorithm becomes, after iterations, there result stored somewhere ? because if think programmer, if re-run program, must store previous results somewhere, or overwritten ? or need use array example store results.

for example, if train image recognition algorithme bunch of cats pictures data sets, variables need add algorithme, if use image library, success everytime find cat, use what? since there nothing saved next step ?

all videos , tutorials have seen, draw graph decision making visualy, , not applying use in future program ?

for example, example, knn used teach how detect written digit, explicit value use ?

https://github.com/aymericdamien/tensorflow-examples/blob/master/examples/2_basicmodels/nearest_neighbor.py

nb: people clicking on close request or downvoting @ least give reason.

the more there iterations, best our algorithm becomes, after iterations, there result stored somewhere

what you're alluding here optimization part.

however optimize model, first have represent it.

for example, if i'm creating simple linear model predict house prices using surface in square meters might go model:

price = * surface + b 

that's representation.

now have represented model, want optimize it, i.e. find params a , b minimize prediction error.

there result stored somewhere ?

in above, have learned params or weights a , b.

that's keep, weights come optimization (also called training) , of course model itself.