ResNetInterface.load_model_file - unsure what I've done wrong
I've worked through the tutorials, and I'm now trying to apply the steps to my own data in the attempt to make a detector for 20 Hz fin whale calls. I'm currently stuck on the following (in the North Atlantic Right Whale detector - part 2 tutorial):
model, audio_repr = ResNetInterface.load_model_file(model_file='fw20.kt', new_model_folder='./fw20_tmp_folder', load_audio_repr=True)
And I get the error message:
`FileNotFoundError Traceback (most recent call last) in 1 #classifier = ResNetInterface.load_model_file('fw20.kt') ----> 2 model, spec_config = ResNetInterface.load_model_file(model_file='fw20.kt', new_model_folder='./fw20_tmp_folder', load_audio_repr=True)
~/anaconda3/envs/ketos_env/lib/python3.6/site-packages/ketos/neural_networks/dev_utils/nn_interface.py in load_model_file(cls, model_file, new_model_folder, overwrite, load_audio_repr, replace_top, diff_n_classes) 731 if load_audio_repr is True: 732 audio_repr = [] --> 733 f = open(os.path.join(new_model_folder,"audio_repr.json"), 'r') 734 json_content = json.load(f) 735 for section, rep in json_content.items():
FileNotFoundError: [Errno 2] No such file or directory: './fw20_tmp_folder/audio_repr.json'`
Going from the comparison to the tutorial script, I'm wondering if there's anything wrong with the model file 'fw20.kt'. I checked the documentation for more details on ResNetInterface.load_model_file, but I can't see what exactly the 'load_audio_repr' argument does.
If someone could point me in the right direction I'd be very grateful! Thanks