Home › Glossary › Systems › Input generator

Intermediate · Systems

Input generator

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. A component that processes raw data into the tensor format suitable for input into a neural network during training or inference.

Technical Definition

A component that processes raw data into the tensor format suitable for input into a neural network during training or inference.

How it works

An input generator is a utility within a machine learning pipeline responsible for data loading and preprocessing. It takes raw data, transforms it into tensors, and then yields batches of these tensors that can be efficiently fed into a neural network. This component is essential for managing the flow of data during training, evaluation, and inference phases.

Related Concepts

  • Neural Network — A computing system inspired by biological neural networks that learns patterns from data through interconnected layers of nodes.
  • Training — The process of adjusting a model's parameters so it learns patterns from labeled or unlabeled data.
  • Data Preprocessing — Transforming raw data into a format suitable for model training.
  • Tensor — A multidimensional array used to represent data in many dimensions, fundamental to deep learning.

Further Reading

  • Google ML Glossary