liveBook · Manning

Signal

As explained in Vibe Engineering

A feature is an input variable or observed attribute supplied to a machine-learning model for making a prediction.

A feature is an input variable—a clue supplied to a model to help make a prediction. The value being predicted is the label: for price prediction, features provide clues and price is the answer. section 4.5.1

Features can take different forms:

  • For drug-price prediction, examples include dosage, salt name, and manufacturer. section 4.5.1
  • For risk classification, medicine descriptions and side effects are combined into one text feature. section 4.6.3

In supervised learning, feature columns form X and labels form y; their rows must stay aligned. A trained model receives test-set features to predict labels, which are compared with the held-out actual values. section 4.5.1 section 4.5.3

Table 4.3 Difference between X and Y datasets
Term
Traditional Name
Math Symbol
Shape
Description
Features
Independent Variables
X
Matrix (N rows * M columns)
The "clues" (Dosage, Salt Name, etc.)
Label
Dependent Variable
y
Vector (N rows * 1 column)
The "answer" you want (Price)

Compare the feature matrix X with the label vector y. section 4.5.1

How Vibe Engineering covers it

How deep it goes
Defines features as model clues and describes their columns in X, without detailing feature selection or transformation.
Examples
Uses dosage, salt name, and manufacturer for price prediction, and combines medicine descriptions with side effects for risk classification.
What you can do after
Apply features as X inputs to train a model and predict labels from test data.
How it connects
Connects features to labels, supervised learning, and train/test splits.

Also explained in

Compare how 14 books explain signal →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Vibe Engineering. The section links point to the parts of the book it draws on.