Index Ai Academy
Foundation · Numbers

Percentages

Per hundred, percent change, percentage points, and the accuracy and error rates of models.

About 35 minutes · Free lesson

What it covers6

Why it matters

Percentages let you compare shares of different totals by putting them all out of 100. They're also how almost every result in machine learning is reported: accuracy, error rate, how much better a new model is. This lesson covers the three forms of one number, percent of an amount, percentage change, and the difference between a change in percent and a change in percentage points, which trips up headlines and reports alike.

What you'll be able to do

  • Move between fractions, decimals and percentages, and find a percent of an amount
  • Work out a percentage change from the old value, and tell it apart from percentage points
  • Write percentages as decimals in Python, and show them with :.1%
  • Report a model's accuracy and error rate, and compare two models honestly
Quick check
Have a guess before you start. A shop cuts a price by 50%, then raises the new price by 50%. Compared with where it started, the price is now…
  1. Back where it started
  2. Higher than at the start
  3. Lower than at the start
  4. It depends on the price
Show the answer

Lower than at the start

From 100, a 50% cut gives 50. A 50% rise on 50 adds 25, giving 75. The second 50% was of a smaller amount, so the price ends lower.

The math: per hundred

Purpose. Which is the bigger share: 3 out of 20, or 14 out of 100? Different totals make shares hard to compare. A percentage rewrites every share out of the same total, 100, so they line up.

Percent means per hundred. 35% is 35 per hundred: 35/100, or 0.35. So a fraction, a decimal and a percentage are three ways to write one number:

fractiondecimalpercentage
1/20.550%
3/40.7575%
3/200.1515%

To go from a decimal to a percentage, times by 100. To go back, divide by 100. So 3 out of 20 is 15%, which beats 14 out of 100.

Quick check
From the last lesson: in 0.07, the 7 is in which place?
  1. Tenths, so 0.07 is 70%
  2. Thousandths, so 0.07 is 0.7%
  3. Ones, so 0.07 is 7
  4. Hundredths, so 0.07 is 7%
Show the answer

Hundredths, so 0.07 is 7%

The first place after the point is tenths, the second is hundredths. 0.07 is 7 hundredths, and 7 per hundred is 7%.

A percent of an amount. To find 20% of 50, turn the percentage into a decimal and multiply: 0.2 × 50 = 10. "Of" means multiply, just as it did with fractions.

Percentage change compares a change with where you started:

change %=new−oldold×100\text{change \%} = \frac{\text{new} - \text{old}}{\text{old}} \times 100

In plain words: find how much it changed, divide by where it started, then times 100.

Watch it work for a club that grew from 40 members to 50. The two bars show the old and new sizes. The picture above them cuts the old 40 into 100 equal squares, so the percentage is how many squares the change fills.

Try it now

Step through, then set the old value to 50 and the new value to 40 and press Use these numbers: the same 10 members, a different percentage, because the hundred-square is now cut from 50.

Percentage points. When both numbers are already percentages, a change can be described two ways. Say a pass rate goes from 40% to 50%. The gap is 10 percentage points: plain subtraction on the 0 to 100 scale. The percentage change is 10 ÷ 40 × 100 = 25%. Both are true, but they answer different questions. A careful report says which one it means.

Now the opening question. Cutting 100 by 50% gives 50. Raising 50 by 50% adds half of 50, which is 25. That gives 75. The second 50% was of a smaller amount, so the price ends lower than it started.

Quick check
A pass rate rises from 20% to 30%. Which statement is right?
  1. It rose 30%, from 20 to 30 points
  2. It rose 10 points, a 50% rise
  3. It rose 50 points, a 10% rise
  4. It rose 10%, which is 10 points
Show the answer

It rose 10 points, a 50% rise

The gap on the 0 to 100 scale is 30 − 20 = 10 points. Compared with the start, 10 ÷ 20 × 100 = 50%.

Shortcuts for checking. 10% of anything is a tenth: move the point one place left, so 10% of 80 is 8. From there, 5% is half of 10%, 20% is double it, and 1% is a tenth of it. So 15% of 80 is 8 + 4 = 12. And 35% of 60 is three lots of 6, plus 3, which makes 21. These are worth knowing even with Python on hand. They tell you at a glance whether a printed answer is the right size.

In your own words
In your own words: why does a 50% cut followed by a 50% rise leave a price lower than it started?

Write a sentence or two, then compare it with a good answer in the lesson.

In Python: percentages are decimals in disguise

Purpose. In Python, the % sign means remainder, as you saw in lesson 2. So code holds a percentage as a decimal, and shows it as a percentage only when it's printed.

share = 3 / 20
print(share)
print(share * 100)
print(f"{share * 100:.1f}%")
print(f"{share:.1%}")
It runs right here in your browser, with nothing to install.

The last line uses a format made for this: :.1% times the number by 100, shows 1 decimal place, and adds the sign. A percent of an amount is the percentage divided by 100, times the amount:

price = 50
discount = 20 / 100 * price
print(discount, price - discount)

A percentage change is three short lines, the same steps as the flow:

old = 40
new = 50
change = new - old
print(change / old * 100)

A named mistake: multiplying by the number in front of %. 20% of 50 isn't 20 × 50. That gives 1000, twenty times the whole price. The 20 means 20 hundredths, so divide it by 100 first:

print(20 * 50)          # wrong: 20 times the price
print(20 / 100 * 50)    # right: 20 hundredths of the price

A quick sense check catches it: any percent under 100 of an amount must give less than the amount.

Worked example

Worked example: a month of growth

A reading app had 1,250 users in May and 1,400 in June. Work out the change, the percentage change, and 15% of June's users.

may = 1250
june = 1400
change = june - may
pct = change / may * 100
print("change:", change)
print(f"percentage change: {pct:.1f}%")
print("15% of June:", 15 / 100 * june)
In the lesson you can change this and run it.

The change is 150 users. Divided by May's 1,250, the starting value, that's 0.12, and times 100 gives 12.0%. The last line turns 15% into 15/100 before multiplying, so it gives 210 users, well under the 1,400 total, which passes the sense check. Inside the print, "15%" is just text in quotes, so Python doesn't treat it as a remainder.

Quick check
What does this print?
score = 0.875
print(f"{score * 100:.1f}%", round(score * 100))
  1. 87.5% 87
  2. 0.875% 1
  3. 87.5% 88
  4. 87.5 88
Show the answer

87.5% 88

score × 100 is 87.5, shown to 1 place with a % added. round(87.5) is a tie, and Python rounds ties to the even neighbor, 88.

Quick check
Which line finds 8% of total?
  1. 8 * total
  2. 8 / 100 * total
  3. total / 8
  4. 8 % total
Show the answer

8 / 100 * total

8% is 8 hundredths, so divide by 100 and multiply by the total.

Changes as multipliers. A 20% rise keeps all of the price and adds 0.2 of it, so it's the same as times 1.2. A 20% cut keeps 0.8 of it, so it's times 0.8. Step through the code below and watch each percentage turn into a plain multiplier before it touches the price.

This view makes the opening question easy. A 50% cut is times 0.5 and a 50% rise is times 1.5, so the two together are times 0.5 × 1.5 = 0.75: the price ends at 75% of where it started, whatever it was. It also shows how to undo a change: divide by the multiplier.

print(0.5 * 1.5)       # a 50% cut, then a 50% rise
up = 50 * 1.2          # a 20% rise
print(up * 0.8)        # a 20% cut doesn't undo it
print(up / 1.2)        # dividing by 1.2 does
Drill

Three everyday percentages

Set as_pct to part out of whole as a percentage, tip to 15% of bill, and rise to the percentage change from old to new. Each starter line has one slip.

part = 18
whole = 24
bill = 60
old = 120
new = 150
as_pct = part / whole
tip = 15 * bill
rise = (new - old) / new * 100
print(as_pct, tip, rise)
In the lesson, your code is checked as you write it.
In your own words
Why does Python code write percentages as decimals?

Write a sentence or two, then compare it with a good answer in the lesson.

In AI and ML: accuracy, error rate and points

Purpose. Almost every result in machine learning is a percentage: how often the model is right, how often it's wrong, and how much better a new model is.

The most common metric is accuracy: the share of examples the model got right. If the sign reader gets 912 of 1,000 test signs right, its accuracy is 91.2%. The flip side is the error rate, the share it got wrong:

error rate=1−accuracy\text{error rate} = 1 - \text{accuracy}

In plain words: whatever share it didn't get right, it got wrong. 91.2% right means 8.8% wrong.

right = 912
total = 1000
accuracy = right / total
print(f"accuracy {accuracy:.1%}, error rate {1 - accuracy:.1%}")
Quick check
What does this print?
acc = 0.95
print(round(1 - acc, 2))
  1. 0.5
  2. 0.95
  3. 0.05
  4. 5
Show the answer

0.05

1 − 0.95 is a hair off 0.05 in floats, and rounding to 2 places tidies it to 0.05: a 5% error rate.

In the real tools. In PyTorch, the library most models are built with, the model's answers and the right answers are kept as tensors, and accuracy is one line. Here are eight test signs, with each kind of sign stored as a number (0, 1 or 2); the reader gets seven of them right:

Points versus percent, in ML. A new sign reader goes from 90% to 95% accurate. That's 5 percentage points, which sounds small. But look at the errors: the error rate falls from 10% to 5%, so the model now makes half as many mistakes. For a reader of a million signs a day, that's 50,000 fewer wrong signs.

old_err = 1 - 0.90
new_err = 1 - 0.95
print(round(old_err, 2), round(new_err, 2))
print(f"{(new_err - old_err) / old_err:.0%} change in errors")

When accuracy is already high, comparing error rates shows progress more honestly than points of accuracy do.

Quick check
Model A is 98% accurate and model B is 99%. How do their errors compare?
  1. They're about the same, 1 point apart
  2. B makes 99% fewer errors than A
  3. B makes 1% fewer errors than A
  4. B makes half as many errors as A
Show the answer

B makes half as many errors as A

A's error rate is 2% and B's is 1%. Going from 2% to 1% halves the errors, a 50% fall, though accuracy rose only 1 point.

Drill

Compare two sign readers

Two sign readers were tested on 2,000 signs. Model A got 1,840 right and model B got 1,920. Set acc_a and acc_b to their accuracies as decimals, points to the gap in percentage points, and fewer to the percentage fall in errors from A to B, as a positive number.

right_a = 1840
right_b = 1920
total = 2000
acc_a = right_a
acc_b = right_b
points = acc_b - acc_a
fewer = 0
print(acc_a, acc_b, points, fewer)
In the lesson, your code is checked as you write it.
In your own words
Why can error rates show progress better than accuracy when a model is already good?

Write a sentence or two, then compare it with a good answer in the lesson.

Goes deeper in

Practice what you just read

  • 8 code examples you can edit and run
  • a practice exercise that checks your code
  • a 5-question quiz
  • 9 review cards for spaced repetition

Try a quiz question

What is 7/20 as a percentage?

  1. 35%
  2. 7.2%
  3. 27%
  4. 0.35%

Answer it at the end of the lesson to see if you're right, and why.

Before this lesson

Keep learning

Where this lesson leads.

575 lessons in thirteen phases. 176 are free without an account, and a free account opens the rest. See every lesson.

Foundation · about 96 hours of study

Foundation math

From whole numbers to the math inside a neural network: each idea by hand, in Python, and in AI 118 lessons All free
  1. Numbers and place value FreeWhat a digit's place is worth, the number line, and Python as a calculator that never tires.36 min
  2. Multiplying and dividing: groups FreeEqual groups, quotient and remainder, and cutting training data into batches.35 min
  3. Order of operations, and names for numbers FreeBrackets, powers, then multiply and divide, then add and subtract; variables hold numbers.36 min
  4. Numbers below zero FreeThe number line through zero, sign rules, distance from zero, and an error's direction and size.37 min
  5. Fractions: equal parts, equal fractions, adding FreePieces of a whole, equal fractions, and re-cutting to add; accuracy as a fraction.45 min
  6. Multiplying and dividing fractions FreeA share of a share, dividing by flipping, and why chances in a row get small.35 min
  7. Decimals and rounding FreeFractions in place value, rounding, and why 0.1 + 0.2 isn't exactly 0.3.36 min
  8. Percentages FreePer hundred, percent change, percentage points, and the accuracy and error rates of models.35 min
  9. Ratios, rates and proportion FreeSharing in a ratio, rates per unit, scaling, and splitting data 80/10/10.35 min
  10. Powers and roots FreeRepeated multiplying, the laws of exponents, square roots, and squared error.43 min
  11. Very big and very small numbers FreePowers of ten, e-notation like 3e-4, orders of magnitude, and a model's size in memory.36 min
  12. Bits, bytes and floats: how numbers are stored FreeBase 2, what n bits can hold, a float as sign, power and digits, and a tensor's memory.35 min
  13. Review: find the slip (numbers) FreeBroken calculations to fix, mixed practice from the whole module, and its formula sheet.34 min
  14. Letters for numbers FreeVariables, terms and like terms, substitution, and a model's formula as a Python function.41 min
  15. Expanding and factoring FreeBrackets as areas, pulling out common factors, and a squared error expanded into a bowl.41 min
  16. Solving equations: keep the balance FreeThe same move on both sides, undoing steps in reverse, and running a model's formula backward.41 min
  17. Rearranging a formula FreeMaking any letter the subject, round-trip checks, and planning a training run.36 min
  18. Inequalities and true-or-false FreeRanges, flipping the sign, if and elif, and the thresholds, clips and ReLU inside a model.42 min
  19. Functions: a rule from input to output Freef(x) notation, tables and graphs, domain and range, loops over lists, and loss as a function.42 min
  20. Chaining and undoing functions Freeg(f(x)), inverses that undo, scaling and unscaling, and a network as a chain.42 min
  21. Straight lines: slope and intercept FreeRise over run, the line through two points, and a model's weight and bias as slope and intercept.42 min
  22. Two equations, two unknowns FreeSubstitution and elimination, where lines cross, and why models are fitted, not solved.42 min
  23. Quadratics: bowls and their lowest point FreeRoots by factoring and by formula, the discriminant, the vertex, and the best weight of a model.42 min
  24. Exponentials: growth and decay FreeThe same factor each step, the number e, and decaying learning rates.43 min
  25. Logarithms: the power question FreeTo what power?, the laws of logs, and adding log-probabilities instead of multiplying.45 min
  26. Sequences and Σ FreeArithmetic and geometric sequences, Σ as a loop that adds, and mean squared error.40 min
  27. Weighted and moving averages FreeWeights that add to 1, a window average, the exponential moving average, and smoothing a loss curve.35 min
  28. Sets, and reading the symbols FreeSets, union and intersection, ℝⁿ and f : ℝⁿ → ℝᵐ, and reading a paper's first equation.41 min
  29. Review: find the slip (algebra) FreeBroken algebra and code to fix, mixed practice from the whole module, and its formula sheet.34 min
  30. Transposing step by step: undo in reverse order FreeA formula's layers, the opposite of each, peeling from the outside in, conditions, and round-trip checks.35 min
  31. Formulas with fractions: clear them first FreeClearing denominators, cross-multiplying, a letter in the bottom, and reciprocal formulas like 1/R = 1/R₁ + 1/R₂.35 min
  32. When the letter appears twice: collect and factor FreeGathering the letter on one side, factoring it out, dividing by the bracket, and rearranging precision and F1.35 min
  33. Squares, roots and powers FreeSquare roots with ±, squaring to undo a root, cube and nth roots, and the √d inside attention.40 min
  34. When the letter is squared and plain: quadratics FreeRearranging into ax² + bx + c = 0, the formula, completing the square, and choosing the root that fits.35 min
  35. Letters in the power: exponentials and logarithms FreeTaking logs to bring a power down, exponentiating to free a letter, and the logit that undoes the sigmoid.40 min
  36. Rearranging inequalities: when the sign flips FreeThe four moves on an inequality, negatives and reciprocals that flip it, logs below 1, and sample sizes.35 min
  37. Inverse functions, lines and scaling FreeUndoing a function by making x the subject, domains that allow it, line forms, and unscaling predictions.40 min
  38. Sequences and sums: find n, a, d and r FreeEvery letter of the arithmetic and geometric formulas as the subject, and an average's window from its decay.40 min
  39. Combining formulas: substitute and eliminate FreePutting one formula inside another to remove a letter, and sizing a training run from a time or money budget.35 min
  40. Review: rearrange anything (algebra) FreeBroken rearrangements to fix, mixed practice from the whole algebra block, a strategy chart, and its formula sheet.35 min
  41. Coordinates and distance FreePoints as (x, y), Pythagoras, distance in any number of coordinates, and nearest neighbors.42 min
  42. Angles, degrees and radians FreeMeasuring a turn, converting units, arcs, and the angle between two arrows.36 min
  43. Sine and cosine FreeAcross and up: right triangles, the unit circle, and directions and rotations.42 min
  44. Waves: height, speed and shift FreeAmplitude, period and phase, sampling a wave, and position codes in transformers.37 min
  45. Review: find the slip (shapes and waves) FreeBroken geometry and code to fix, mixed practice from the module, and its formula sheet.29 min
  46. Angles, triangles and waves: rearranging with sin, cos and tan FreeSides and angles from SOH CAH TOA, principal values, arcs, polar coordinates, and a wave's time and period.45 min
  47. Vectors: lists of numbers that point FreeA list, a point and an arrow; components, dimension, NumPy arrays, and data as vectors.42 min
  48. Adding and scaling vectors FreeComponent-wise sums, scalar multiples, mixes, and the weight update w − ηg.43 min
  49. Length and distance: norms FreeL2, L1 and max norms, unit vectors, distance, and the norms inside losses and clipping.42 min
  50. The dot product FreeMultiply in pairs and add; a·a as squared length; a @ b; one neuron as a dot product.43 min
  51. Angle, cosine similarity and projection FreeDirection without length, perpendicular vectors, shadows, and search by meaning.45 min
  52. Perpendicular directions: bases and coordinates FreeOrthonormal directions, coordinates as dot products, rebuilding a vector, and taking a direction out.35 min
  53. Matrices: tables of numbers FreeRows, columns and entries; two-dimensional arrays and slices; a data set as a matrix.41 min
  54. Adding, scaling and transposing matrices FreeEntry-by-entry sums and scalings, the transpose, mean images, and tables turned around.36 min
  55. A matrix times a vector FreeEach row dotted with the vector, the size rule, A @ x, and a layer acting on one example.42 min
  56. Matrix multiplication FreeRow meets column, the size rule, order matters, and a batch through a layer: X W + b.45 min
  57. Identity, diagonal, triangular and symmetric FreeFour patterns and their jobs: rescaling, distance tables, and the causal mask.37 min
  58. Matrices as movements of space FreeStretches, flips, shears and rotations; columns as landing spots; chaining; a layer moving data.42 min
  59. Determinant and inverse FreeThe area factor ad − bc, singular matrices, undoing a matrix, and redundant features.42 min
  60. Solving systems by elimination FreeAugmented matrices, row moves, back-substitution, the three outcomes, and fitting weights.45 min
  61. Independence, span and rank FreeRedundant vectors, span, rank, outer products, and low-rank updates like LoRA.42 min
  62. Eigenvectors: directions that only stretch FreeAv = λv, checking candidates, np.linalg.eig, and the main direction of a data cloud.41 min
  63. SVD and PCA: the important directions first FreeA matrix as pieces by size, the best rank-k copy, compression, and PCA maps of data.45 min
  64. Tensors, shapes and broadcasting FreeAxes and shapes, reshape, axis=, broadcasting, and shapes through a small model.43 min
  65. One-hot vectors and lookup tables: embeddings FreeA one-hot row picks a table's row, a sentence looked up as vectors, and a score for every word.35 min
  66. Sliding dot products: convolution FreeA small pattern slid along a list and over a grid, output sizes, stride and padding, and shared weights.40 min
  67. Review: find the slip (vectors and matrices) FreeShape slips to fix, mixed practice from the module, and its formula sheet with shapes.36 min
  68. Vector equations: rearranging with arrows FreeSolving for a vector or a scale, recovering a gradient from two weight snapshots, and an angle from a dot product.40 min
  69. Matrix equations: multiply by the inverse, on the correct side FreeNo dividing by a matrix: inverses from the left or right, order, transposes, and the normal equations.40 min
  70. Rates of change FreeChange over change, secants, rates from readings with np.diff, and reading a loss curve.41 min
  71. Limits: what a value heads toward FreeApproaching from both sides, holes, corners and jumps, epsilon, and judging activations.37 min
  72. The derivative: slope at a point FreeShrinking secants, f′(x) and dy/dx, Δy ≈ f′Δx, and which way a weight should move.45 min
  73. Derivative rules FreeConstants, powers, multiples, sums, eˣ and ln x, checked numerically, and a squared error's slope.41 min
  74. The product rule and the chain rule FreeProducts and chains of functions, forward then backward, and one neuron's dL/dw.45 min
  75. Lowest points and curvature Freef′ = 0, the second-derivative test, tables, local and global minima, and convex losses.43 min
  76. Standing in for a curve near a point FreeTangent lines, a curvature term, growing error, and why learning rates must be small.37 min
  77. Partial derivatives FreeOne input at a time, ∂, nudging in Python, and the two slopes of a line's loss.41 min
  78. The gradient FreeAll the slopes as a vector, steepest climb and descent, a numerical gradient, and the loss gradient.42 min
  79. Gradient descent FreeStepping against the gradient, the learning rate, a training loop, and fitting a line.45 min
  80. Faster descent: momentum and Adam FreeMomentum and Adam by hand, the bias correction, and three optimizers down one narrow valley.40 min
  81. The chain rule with many paths: backpropagation by hand FreeGraphs, multiply along paths and add across them, gradient checks, and a 2-2-1 network by hand.45 min
  82. Automatic differentiation: how PyTorch finds every slope FreeA tape of steps walked backward, and requires_grad, backward(), .grad, zero_grad() and no_grad().40 min
  83. Jacobians and Hessians FreeSlopes of vector functions, a layer's Jacobian, second partials, and bowls, domes and saddles.42 min
  84. Integrals: adding up thin slices FreeStrip and trapezoid sums, summing undoes differencing, probability as area, and AUC.43 min
  85. Review: find the slip (calculus) FreeCalculus slips to fix, mixed practice from the module, and its formula sheet.36 min
  86. Rearranging in calculus: slopes, steps and constants FreeSetting a slope to zero, Newton's step, the chain rule turned around, constants from a point, and learning rates.45 min
  87. Counting: choices, orders and groups FreeThe counting principle, factorials, ordered picks and groups, and why attention's cost grows with the square of the length.43 min
  88. Probability: from counts to chances FreeEqually likely outcomes, the complement, distributions, seeded simulation, and a next-word model.43 min
  89. Combining events: and, or, independent FreeThe or rule, independence, probability trees, drawing with and without replacement, and chains of steps.42 min
  90. Conditional probability and Bayes' rule FreeConditionals from tables, why P(A | B) isn't P(B | A), Bayes' rule, and precision and recall.45 min
  91. Random variables and distributions FreeUncertain numbers, the probability of a range, sampling by slices, and sampling versus argmax.42 min
  92. Expected value, variance and standard deviation FreeProbability-weighted averages, spread in the data's own units, and a training loss as an expected value.42 min
  93. Bernoulli, binomial, uniform and normal FreeYes-or-no trials, counts of yeses, the 68-95-99.7 rule, z-scores, and an accuracy's error bar.45 min
  94. Two variables at once: covariance and correlation FreeJoint tables and marginals, covariance, correlation, the covariance matrix, and features that carry the same information.45 min
  95. Review: find the slip (probability) FreeProbability slips to fix, mixed practice from the module, and its formula sheet.36 min
  96. Probability formulas: solve for the missing chance FreeConditionals, Bayes' rule for the prior, the or rule, expected values, variance, and z-scores turned around.45 min
  97. Describing data: center, spread and scale FreeMean, median and mode, variance and standard deviation, percentiles, outliers, and standardizing features.42 min
  98. Samples, the law of large numbers and the central limit theorem FreeSamples and populations, the standard error σ/√n, the bell of averages, and error bars on test scores.43 min
  99. Is the difference real? Comparing two models FreePaired differences, a bootstrap interval, what a p-value is and isn't, and why many tries make lucky wins.40 min
  100. Maximum likelihood FreeLikelihoods as products, log-likelihoods as sums, grid search, underflow, and the negative log-likelihood loss.45 min
  101. Linear regression: the best line FreeResiduals, least squares by formula, three roads to one line, the matrix form, and train versus test error.45 min
  102. Logistic regression: a line that answers yes or no FreeThe sigmoid, the log loss, its slope p − y, a training loop, and choosing a threshold.45 min
  103. Scoring a classifier: confusion table, precision, recall, F1 FreeThe four cells, accuracy, precision, recall and F1, and what moving a threshold does to each.35 min
  104. Softmax: scores into probabilities FreeExponentiate and share out, shifting and temperature, the safe softmax, and a classifier's last step.43 min
  105. Surprise, entropy and cross-entropy FreeSurprise in bits, entropy, cross-entropy and KL divergence, perplexity, and a language model's loss.45 min
  106. KL divergence: how far one distribution is from another FreeKL by hand, cross-entropy as entropy plus KL, why it isn't symmetric, and a soft-target loss.35 min
  107. Review: find the slip (statistics and information) FreeStatistics slips to fix, mixed practice from the module, and its formula sheet.38 min
  108. Statistics and information: n, z, logits and bits FreeSample sizes from error bars, inputs from probabilities, logit gaps from softmax, and chances from bits and losses.40 min
  109. One neuron, start to finish FreeWeighted sum, activation, loss, slopes and update, by hand and in code, and what one straight line can and can't split.45 min
  110. A layer and a small network as matrices FreeTwo layers on a batch, the shape of every matrix, adding a hidden unit, and XOR solved.45 min
  111. Training a small network FreeThe backward pass in matrix form, gradient checks, a full training loop, and reading loss curves.45 min
  112. The same network in PyTorch Freenn.Sequential with the NumPy weights loaded, the same loss and slopes, then SGD and mini-batches.45 min
  113. Attention as matrix math FreeScaled dot-product scores, the causal mask, a row-by-row softmax, and blending value vectors.45 min
  114. Residual connections and layer norm FreeAdding the input back and why its slope keeps learning alive, layer norm by hand, and add and norm.40 min
  115. A transformer block, start to finish FreeA sentence through embeddings, attention, add and norm, a small network and next-word scores, every shape named.45 min
  116. Planning a model: rearranging the formulas of a working network FreeLayer widths from parameter counts, sizes from memory and compute, the attention scale, and a training plan.40 min
  117. Final test: transpose any formula FreeSlips to fix from every topic, mixed practice of every kind of rearranging, the strategy chart and the full sheet.45 min
  118. Capstone: a sign reader from the formulas FreeBuild, train and test a trail-sign classifier with NumPy alone, and report its accuracy with an error bar.45 min
Part of a free course 118 lessons

The Foundation Math For Ai & ML: from arithmetic to attention

Learn the math behind AI from arithmetic up: numbers, algebra, vectors and matrices, calculus, probability and statistics, each shown as numbers moving and checked in Python, ending with a network you build and train from the formulas.

Never trained an AI? Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account. Teach it to write