---
title: "Discussion on \"Splitting Fraud Data Without Breaking the Ratio (Day 3)\" | Hashnode"
description: "Discussion on \"Splitting Fraud Data Without Breaking the Ratio (Day 3)\". Why stratify=y is the most important argument in my train/test split."
url: "https://hashnode.com/posts/splitting-fraud-data-without-breaking-the-ratio-day-3/6abbac6a3973b25f7d9644c3"
---

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

Discussion

# [Splitting Fraud Data Without Breaking the Ratio (Day 3)](https://beginnerslog.hashnode.dev/splitting-fraud-data-without-breaking-the-ratio-day-3?utm_source=hashnode&utm_medium=feed)

Why stratify=y is the most important argument in my train/test split.

Sep 29 · 3 min read · beginnerslog.hashnode.dev  
Read the article

-   [#beginners](https://hashnode.com/tag/beginners)
-   [#scikit-learn](https://hashnode.com/tag/scikit-learn)
-   [#datascience](https://hashnode.com/tag/datascience)
-   [#learninginpublic](https://hashnode.com/tag/learninginpublic)

## Written by

[Faith Mwende](https://hashnode.com/@mwendekiindu)

@mwendekiindu · Nairobi, Kenya

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

Learning in public, one post at a time.

Learning cloud and data science from zero. Writing what I learn so it sticks; the wins, the confusion, and everything in between.

## Responses

Markdown supported · ⌘↵ to post

No responses yet.