For full text search please use the '?' prefix. e.g. ? Onboarding

Terminology

Label

A label is the thing we're predicting—the yy variable in simple linear regression.

  • ex. The label could be the future price of wheat, the kind of animal shown in a picture, the meaning of an audio clip, or just about anything.

Feature

A feature is an input variable—the x variable in simple linear regression. A simple machine learning project might use a single feature, while a more sophisticated machine learning project could use millions of features, specified as:

x1,x2...xNx_{1}, x_{2}...x_{N}
  • ex. In a spam detector, the features could include: words in the email text, sender's address, time of day the email was sent, email contains the phrase "one weird trick."

Example

An example is a particular instance of xx (note: xx is a vector)

  • each example is often represented by a vector of features.

Examples can be either unlabeled or labeled.

  • A labeled example includes both feature(s) and the label
  • An unlabeled example contains features but not the label.

Model

A model defines the relationship between features and label.

  • ex. a spam detection model might associate certain features strongly with "spam".

There are 2 principle phases of a model's life:

  1. Training, which means creating or learning the model. That is, you show the model labeled examples and enable the model to gradually learn the relationships between features and label.
  2. Inference, which means applying the trained model to unlabeled examples. That is, you use the trained model to make useful predictions (y').
    • ex. during inference, you can predict medianHouseValue for new unlabeled examples.

Parameter

A model parameter is a configuration variable that is internal to the model and whose value can be estimated from data.

  • that is, a parameter is a value that the model sets itself

In the context of neural networks, parameters usually refer to weights and biases of the network.

Hyperparameter

A hyperparameter is a value that we (as a developer) set.

  • ex. learning rate

Epoch

Each time a dataset passes through an algorithm, it is said to have completed an epoch

  • therefore, one epoch is one entire passing of training data through the algorithm

An epoch is a hyperparameter that determines the process of training the machine learning model

Backlinks