search

ChangWeiTan / TS-Extrinsic-Regression

This repository contains the source code for time series regression.
GNU General Public License v3.0
93 stars 19 forks source link

[Solved] Encoding ISO-8859-1 was needed to read BIDMC datasets #8

Closed Didayolo closed 10 months ago

Didayolo commented 3 years ago

When trying to read "BIDMC32RR", "BIDMC32HR", "BIDMC32SpO2" datasets, I got the following error:

UnicodeDecodeError: 'utf-8' codec can't decode byte 0x96 in position 1164: invalid start byte

In order to solve it I had to make a change in utils/data_loader.py at line 80:

with open(full_file_path_and_name, 'r', encoding='ISO-8859-1') as file: # NEW
# with open(full_file_path_and_name, 'r', encoding='utf-8') as file: # OLD

I am not sure if the problem is general or came from my system settings. Still seems interesting to share.

ChangWeiTan commented 3 years ago

That is interesting. I never had that issue. Thanks for pointing this out.

javierdjas commented 1 year ago

I had the same issue, it solved it, thank you