Task 6A wrong csv

[asicoderOfficial]

The csv file _clotho_captionsdevelopment.csv contains some entries with trailing spaces, such as [ typical neighborhood in Porto.wav], which should be [typical neighborhood in Porto.wav], without the first space.

To solve it, preprocess the file like:

import pandas as pd

df = pd.read_csv('data/clotho_captions_development.csv')
df['file_name'] = df['file_name'].str.strip()
df.to_csv('data/clotho_captions_development.csv', index=False)

Please upload the csv corrected and close this issue.

[felixgontier]

Hello, After double checking, the 3 audio files in question also contain a leading space, so clothocaptions.csv files do correctly match the audios. For context these names are unchanged from Freesound where they were collected. I agree that this naming is poor practice, and is not always handled well by file systems/libraries. Since it hasn't particularly been a problem for participants in the past, it's unlikely that there will be a complete reupload of the dataset on Zenodo fixing this (note that I am not part of the team that curated the current version, or the owner of the repository). However we'll keep in mind to include it in any future update of Clotho.

[asicoderOfficial]

Thanks for the response! I close the issue, everything clear now.