small fix

new_specie/compute_embeddings.py

@@ -17,8 +17,7 @@ parser.add_argument("-sampleDur", type=float, default=1, help="Size of the signa
 args = parser.parse_args()
 
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
-#frontend = models.frontend(args.SR, args.NFFT, args.sampleDur, args.nMel)
-frontend = models.frontend_gibbon
+frontend = models.frontend(args.SR, args.NFFT, args.sampleDur, args.nMel)
 encoder = models.sparrow_encoder(args.bottleneck // (args.nMel//32 * 4), (args.nMel//32, 4))
 decoder = models.sparrow_decoder(args.bottleneck, (args.nMel//32, 4))
 model = torch.nn.Sequential(frontend, encoder, decoder).to(device)

new_specie/sort_cluster.py

@@ -22,7 +22,6 @@ parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFo
     For insights on how to tune HDBSCAN parameters, read https://hdbscan.readthedocs.io/en/latest/parameter_selection.html""")
 parser.add_argument('encodings', type=str, help='.npy file containing umap projections and their associated index in the detection.pkl table (built using compute_embeddings.py)')
 parser.add_argument('detections', type=str, help=".csv file with detections to be encoded. Columns filename (path of the soundfile) and pos (center of the detection in seconds) are needed")
-#parser.add_argument('audio_folder', type=str, help='Path to the folder with complete audio files')
 parser.add_argument("-audio_folder", type=str, default='./', help="Folder from which to load sound files")
 parser.add_argument("-SR", type=int, default=44100, help="Sample rate of the samples before spectrogram computation")
 parser.add_argument("-nMel", type=int, default=128, help="Number of Mel bands for the spectrogram (either 64 or 128)")