How to Fix Chinese Garbled Text in Jupyter Notebooks: A Step-by-Step Guide

Introduction

In a Python community chat, the author encountered garbled Chinese characters when viewing a Jupyter notebook, while normal console output displayed correctly.

Problem Demonstration

The notebook shows unreadable characters instead of the expected Chinese text.

Attempted Fixes

First, a Python 2‑style solution was tried:

import sys
stdi, stdo, stde = sys.stdin, sys.stdout, sys.stderr
reload(sys)
sys.setdefaultencoding('utf-8')
sys.stdin, sys.stdout, sys.stderr = stdi, stdo, stde

This approach fails under Python 3.

Next, reloading the sys module with importlib.reload was attempted:

import importlib, sys
importlib.reload(sys)

It also did not resolve the issue.

Effective Solution

A contributor suggested setting the Windows console code page to UTF‑8 before launching the notebook:

chcp 65001

Running this command fixes the garbled output.

Why the Error Occurs

The “锟斤铐” characters are a typical symptom of reading UTF‑8 encoded text with a GBK decoder (or vice‑versa). Mismatched encodings cause the unreadable symbols.

Conclusion

The article summarizes the encoding problem, presents several failed attempts, and highlights the simple chcp 65001 fix, helping Python users resolve similar Unicode issues in Jupyter notebooks.