Unexpected: segments.py not extracting all audio clips as expected from .csv (--rtype r)

[GiadaGiacomini]

Hello,

I'm greatly enjoying using BirdNET-Analyzer V.2.4.

When running segments.py on .csv files previously created with analyze.py (--rtype r), the function identify only some of the .csv files (including files that do contain IDs). This is confirmed by reading the csv in R where I could see that segments.py extracted only 875 segments instead of the 1805 expected by the number of lines with IDs in the csv files.

I've tried to run the code on a smaller batch but didn't change the outcome. I've also tried to add a blank row at the end of the .csv files but it didn't change the outcome.

Do you have any suggestions on what I might be doing wrong or on how to fix the issue?

Thanks in advance!
Giada

[Josef-Haupt]

There is indeed a bug with segments.py, because we changed how R tables are generated. I'm going to to fix it soon, hopefully that will also fix your problem.

[GiadaGiacomini]

Thanks Josef for the prompt answer.

Update:
I've ran analyze.py again to get .txt files (--rtype table) but I had the same results as above when running segments.py on the txt files: only some of the .txt files were read by the function and as a result only a part of the segments were extracted.

If the bug is in segments.py, and not in analyze.py, I could continue to run analyze.py on the rest of the data? Otherwise I shall wait.

[Josef-Haupt]

should be fixed