---
title: Normalized Edit Distance
emoji: 📈
colorFrom: blue
colorTo: red
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false
license: apache-2.0
tags:
- evaluate
- metric
short_description: Normalized Edit Distance (NED)
description: >-
  The Normalized Edit Distance (NED) is a metric used to quantify the
  dissimilarity between two sequences, typically strings, by measuring the
  minimum number of editing operations required to transform one sequence into
  the other, normalized by the length of the longer sequence.  The NED ranges
  from 0 to 1, where 0 indicates identical sequences and 1 indicates completely
  dissimilar sequences. It is particularly useful in tasks such as spell
  checking, speech recognition, and OCR. The normalized edit distance can be
  calculated using the formula: NED = (1 - (ED(pred, gt) / max(length(pred),
  length(gt)))) Where: gt: ground-truth sequence pred: predicted sequence ED:
  Edit Distance, the minimum number of editing operations (insertions,
  deletions, substitutions) needed to transform one sequence into the other.
---
# Metric Card for NED

## Metric Description

The Normalized Edit Distance (NED) is a metric used to quantify the dissimilarity between two sequences, typically strings, by measuring the minimum number of editing operations required to transform one sequence into the other, normalized by the length of the longer sequence. 
The NED ranges from 0 to 1, where 0 indicates identical sequences and 1 indicates completely dissimilar sequences.
It is particularly useful in tasks such as spell checking, speech recognition, and OCR. The normalized edit distance can be calculated using the formula:

NED = (1 - (ED(pred, gt) / max(length(pred), length(gt))))
  
  Where:

gt: ground-truth sequence

pred: predicted sequence

ED: Edit Distance, the minimum number of editing operations (insertions, deletions, substitutions) needed to transform one sequence into the other.