AI glossary: terms starting with D

One measured value for each calendar day.

Information in training that the model will not have when it is used; target leakage is one kind.

The findings about a table, each with the decision it leads to: keep, transform, drop, or a question for a later video.

A test for missing, repeated or invalid recorded values before a declared use. It records an issue; the decision about it comes after.

A class whose __init__, __repr__ and __eq__ are written from its list of fields, with @dataclass.

A tool that stops a program so you can look at its values. Python's is pdb; VS Code has one built in.

The predicted label at every point of the feature space.

A function applied to the class or function below it, written with @: @dataclass.

The value a parameter takes when a call does not pass that argument: rate=0.08.

A dictionary that creates a missing key's value when the key is read: defaultdict(list) creates []. From collections.

The highest power in a polynomial model.

all eligible trips in the stated question.

A numerator and denominator that refer to different eligible sets, such as 3,207 approved defaults over all 30,000 rows.

a curve with total area 1, whose area over an interval is the probability of landing there

Example: 2x on [0, 1]

Locations with equal fitted probability density.

The rule mapping estimated density to plotted width. With a separate scale per group, widths do not show group size.

A package a project needs, listed in pyproject.toml.

Running a model on new data as part of a real system.

the slope of the tangent, the limit of Δy/Δx

Example: 2 for x² at 1

a summary of observed data. The median of the January–February trips is one.

The feature table the model is fitted on, one row per sample, X.

the factor a matrix scales every area by; ad − bc for a 2 × 2 matrix

Example: 5

An observed value minus its group's mean.

A collection of key value pairs, looked up by key: {"Pasta": 12.50, "Salad": 8.25}. Its type is dict.

A dictionary built from a key and a value evaluated for each item: {item: price for item, price in zip(items, prices)}.

The values in the first set that are not in the second, a - b.

the number of components

Example: (3, 2, 4) is in three

An axis / the number of axes

Describing examples with fewer columns (Module 11).

the slope of f in the direction of a unit vector u, ∇f · u

Example: 6.4641 at 30°

One of a countable set of values: $0, $100 or $1,000.

The dates included on a chart.

the norm of the difference, ‖a − b‖

Example: 2.2361 from (3, 1) to (1, 2)

in high dimensions, the nearest and farthest neighbours are almost equally far away

Example: ratio 1.123 at d = 1000

Each step landing further from the minimum.

A string on the first line of a function's body that says what the function does. help() and __doc__ show it.

the inputs a function accepts

Example: every real number for 2x + 1; positive numbers for log₂ x

the sum of the products of matching components

Example: (3, 4) · (2, 1) = 10

The type and width used for array elements

Separate vertical scales placed on one plot; either can be rescaled until two lines look aligned.

One 0/1 column for every category, beside an intercept, which makes the columns perfectly collinear.

A row whose content repeats another row.

A row matching an earlier row on the compared fields: 5,268 exact repeats, to inspect, not delete silently.

the same derivative, written as a ratio of small changes

Example: d(x²)/dx = 2x

A name can refer to an object of any type, and can be rebound to an object of a different type: items = 3, then items = 3.5.