---
title: "Discussion on \"When Your Training Data Pipeline Has Three Different Ideas About the Same Thing\" | Hashnode"
description: "Discussion on \"When Your Training Data Pipeline Has Three Different Ideas About the Same Thing\". Silent data loss in ML pipelines, and why normalisation belongs at ingestion"
url: "https://hashnode.com/posts/when-your-training-data-pipeline-has-three-different-ideas-about-the-same-thing/69f852576388755c94d4d65c"
---

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

Discussion

# [When Your Training Data Pipeline Has Three Different Ideas About the Same Thing](https://toddsullivan.hashnode.dev/when-your-training-data-pipeline-has-three-different-ideas-about-the-same-thing?utm_source=hashnode&utm_medium=feed)

Silent data loss in ML pipelines, and why normalisation belongs at ingestion

May 4 · 3 min read · toddsullivan.hashnode.dev  
Read the article

-   [#ai](https://hashnode.com/tag/ai)
-   [#computervision](https://hashnode.com/tag/computervision)
-   [#devops](https://hashnode.com/tag/devops)
-   [#machinelearning](https://hashnode.com/tag/machinelearning)
-   [#python](https://hashnode.com/tag/python)

## Written by

[Todd Sullivan](https://hashnode.com/@toddsullivan)

@toddsullivan

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

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
