Supervised Classification

Calls may be classified with a supervised neural network. This network operates on the spectrogram, rather than the contour.

To use a supervised classifier:

1. Select "Tools > Call Classification > Supervised Classification"

2. In the list box, select the detection files to classify, and click "OK".

To train a supervised classifier:

1. Select "Tools > Network Training > Train Supervised Classifier"

2. Select the detection files to to use for training.

   • Rejected calls and "Noise" labeled calls will be ignored.

3. When training is finished, a dialog will appear to save the classification network

4. Specify the spectrogram frequency range, and click "OK".

5. The network will begin training. A confusion matrix will appear when training finishes, and you will be prompted to save the network.

6. For MATLAB savvy individuals, you can adjust the network parameters in "DeepSqueak/Functions/Network_Training/TrainSupervisedClassifier_Callback.m"