RuntimeError with birdnetlib and Custom Model Analysis

[mohedano-munoz]

Issue Description:

I am encountering a runtime error when attempting to use a custom model trained with birdnet-analyzer and performing analysis with birdnetlib. The specific error message I am receiving is:

RuntimeError: tensorflow/lite/kernels/concatenation.cc:162 t->dims->data[d] != t0->dims->data[d] (1 != 0)Node number 92 (CONCATENATION) failed to prepare.

Python Code:

from birdnetlib.analyzer import Analyzer
from birdnetlib import Recording

model_path = '/path_to_file/test_2.tflite'
labels_path = '/path_to_file/test_2_Labels.txt'

analyzer = Analyzer(classifier_labels_path=labels_path, classifier_model_path=model_path)

recording = Recording(
    analyzer,
    'path_to_audio_file',
    min_conf=0.1
)

recording.analyze()

Environment:

Operating System: Ubuntu 23.10 Python Version: 3.11.5 Birdnetlib Version: 0.12.3

Thank you for your attention to this matter. Let me know if you need further details or clarification.

[joeweiss]

Hi @mohedano-munoz.

Which version of BirdNET-Analyzer was the custom model trained with?

Also, can you confirm that the error does not appear when you use BirdNET-Analyzer's analyze.py with the --classifier argument as outlined here?

Thanks.

[mohedano-munoz]

Hello, the version of BirdNET Analyzer that I'm using is the latest one available on GitHub in its main branch (Revision number: f42e4bf3adc3824b8ab4ec61f151096eb44ebf8f). I have tried running it with BirdNET Analyzer on the same audio files with its version, and it works correctly.

Greetings, and thank you very much for everything.

[joeweiss]

Thanks. Is there any chance I can get those custom model files?

If you'd rather not share them publicly, you can email me at joe.weiss@gmail.com. That would greatly reduce the time it takes to debug this issue. If you need an upload location for the files, I'll be happy to reply with a Dropbox link.

Also, it would help to have the installed libraries for your environment (pip freeze or pip3 freeze).

Thanks!

[joeweiss]

Closing this issue since there hasn't been any response. If you encounter this issue again or have additional information to share, please reopen the issue. Thanks for your understanding!

[mohedano-munoz]

Hi Joe

I'm sorry for not replying here. I sent the generated model through the email address. I have just forwarded the email to you with the model and the environment.

Thank you very much, and greetings!