GitLab

Hi again folks, I'm now running into an issue with the detection portion of the process and am hoping you can help me with the general issue of having 0 detections reported. Here are the issues I'm having:

1. I skipped the spectrogram loading and review portion because I had run into a problem at the "Loading a Classifier" ResNetInterface.load_model_file part at the beginning, and then with the spectrogram not plotting (the command ran and no errors, but no plots), which is likely something related to pycharm.

The line of code I used with this section was: resnet, audio_repr = ResNetInterface.load_model_file(model_file='SOCAL_Mn_Network.kt', new_model_folder='./Mn_test') which followed line 6 of the Creating a detector training module. I did notice that the text below that field mentioned a third argument but I don't see that indicated in the line of code example in the tutorial.

This is the error I got from that step (hope it shows up correctly)

WARNING:tensorflow:A checkpoint was restored (e.g. tf.train.Checkpoint.restore or tf.keras.Model.load_weights) but not all checkpointed values were used. See above for specific issues. Use expect_partial() on the load status object, e.g. tf.train.Checkpoint.restore(...).expect_partial(), to silence these warnings, or use assert_consumed() to make the check explicit. See https://www.tensorflow.org/guide/checkpoint#loading_mechanics for details.
Traceback (most recent call last):
  File "C:\Users\eferg\.conda\envs\ketos_env\lib\site-packages\IPython\core\interactiveshell.py", line 3444, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-35-4b79115ec4b4>", line 1, in <module>
    resnet, audio_repr = ResNetInterface.load_model_file(model_file='SOCAL_Mn_Network.kt', new_model_folder='./Mn_test')
TypeError: cannot unpack non-iterable ResNetInterface object

2. When I skipped down to the batch processing, I'm able to execute all the lines of codes similar to in the tutorial, but when I get to the print detections section, in indicates there are not detected calls (and the progress bar goes very quickly). I'm guessing since the model worked fine-ish with the training and validation data that it should be detecting the calls in the file.

I have a link to the wav file below, along with the network and code (starting around line 247) below. Appreciate any thoughts!

Wav and model files SOCAL_Mn_Network.kt BOEM_Mn_Network.py