Lists and loops over real data
Why it matters A neural network is mostly loops over lists of numbers. If you can total, filter and count a list, you can read most training code.
Say you logged your steps for a week: [4200, 8100, 12000, 3000, 9500, 10100, 7600]. Two questions you might ask: how many steps in total, and on how many days did you reach a goal of 8,000?
See it
A loop visits the items one at a time. You keep a running number and update it on each visit, like a tally on paper.
total = 0
for s in steps: # s is 4200, then 8100, then 12000 ...
total = total + s
After the first visit total is 4200, after the second 12300, and so on.
Running total: a variable you start at zero and add to on every pass through the loop.
Counting is the same idea, except you add 1 only when a condition is true: if s >= goal: count = count + 1.
Recap Start a number, loop, update it. That pattern is how a network adds up its loss over a dataset.
Write two functions:
total_steps(steps)returns the sum of the list.days_at_goal(steps, goal)returns how many days are at or abovegoal.
Use a loop for both (no sum()), so you practise the pattern.
Spotted a problem in this lesson? Report it · Ask about this lesson
This exercise opens soon
Finish the lesson before it and this one unlocks. The whole course is free, no account needed, and you can read this lesson right now.
This browser is blocking cookies for learn.floati.life, so lessons you finished here cannot open in order. Allow cookies for this site and reload.