---
title: "Discussion on \"Why 0.5 Isn't Always the Right Classification Threshold\" | Hashnode"
description: "Discussion on \"Why 0.5 Isn't Always the Right Classification Threshold\". I changed the threshold of the same churn model from 0.3 to 0.7. The model stayed the same, but its predictions changed quite a lot."
url: "https://hashnode.com/posts/why-0-5-isn-t-always-the-right-classification-threshold/6af2ffa99538fc7c0f2ddfef"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Why 0.5 Isn't Always the Right Classification Threshold](https://vashniroy.hashnode.dev/why-0-5-isn-t-always-the-right-classification-threshold?utm_source=hashnode&utm_medium=feed)

I changed the threshold of the same churn model from 0.3 to 0.7. The model stayed the same, but its predictions changed quite a lot.

5h ago · 3 min read · vashniroy.hashnode.dev  
Read the article

-   [#machine-learning](https://hashnode.com/tag/machine-learning)
-   [#classification](https://hashnode.com/tag/classification)
-   [#model-evaluation](https://hashnode.com/tag/model-evaluation)
-   [#python](https://hashnode.com/tag/python)
-   [#data-science](https://hashnode.com/tag/data-science)

## Written by

[Vashni Roy](https://hashnode.com/@vashniroy)

@vashniroy · Hyderabad, India

  
[View profile](https://hashnode.com/@vashniroy)

Software Engineer | Full Stack & AI/ML

Software engineer working across full-stack development and AI/ML. I write about experiments, engineering decisions, and things I find useful while building software.

## Responses

Markdown supported · ⌘↵ to post

No responses yet.