unicodedecodeerror 'utf-8' codec can't decode byte 0xb4 in position 29 invalid start byte because of "한글"
2023, Oct 31
Today, during a machine learning lecture, there was a significant time thief that disrupted my progress. It was the error message
unicodedecodeerror: 'utf-8' codec can't decode byte 0xb4 in position 29: invalid start byte.
Although I searched for solutions within the lecture’s Q&A, only TensorFlow and NumPy version issues were mentioned. I even reinstalled Anaconda3 and tried changing versions, but in the end, I found the solution through a Google search.
The error was caused by the presence of Korean characters in the project folder’s path. When I moved my project to the root folder of the C drive, it started working perfectly. I guess I’ll have to remove all Korean characters from my computer setup in the future.