Phase 2 · Math, when you need it · Calculus
Functions and graphs
Read a graph and see how an output depends on an input.
Open this lesson Create a free account
Why it matters
A trained model is a function: text in, probabilities out. Its loss is a function of its weights, and training is the search for the lowest point on that function's graph. Marketing is full of functions too: revenue as a function of price, conversions as a function of ad spend. Being able to look at a graph and say "this flattens out", "this has a peak here" is the intuition the next four lessons (derivatives) make precise.
What you'll be able to do
- Read function notation and evaluate functions on whole arrays of inputs
- Read a graph: where it rises, falls, flattens and peaks
- Recognize the shapes that recur in machine learning: line, bowl, S-curve, hinge
- Find the best input for a function by evaluating many candidates
Before this lesson
Next in this phase: Exponentials and logarithms