def train_and_test_network():
"""
Train a neural network and test it. Can also train on other feature types,
or run the experimenter to run different configurations
"""
min_speakers = 1
max_speakers = 10
# Load data from filesystem
data_loader = DataLoader(train_dir, test_src_dr, test_dest_dir)
data_loader.force_recreate = False
data_loader.min_speakers = min_speakers
data_loader.max_speakers = max_speakers
# Train network
train, (test_x, test_y) = data_loader.load_data()
libri_x, libri_y = data_loader.load_libricount(libri_dir)
# Train and test network
file = 'testing_rnn'
net = RNN()
net.save_to_file(file)
net.train(train, min_speakers, max_speakers, FEATURE_TYPE)
net.load_from_file(file)
timit_results = net.test(test_x, test_y, FEATURE_TYPE)
libri_results = net.test(libri_x, libri_y, FEATURE_TYPE)
def test_networks(self):
"""
Test the networks saved by run(). Save to json
"""
data = self.__get_test_data()
result = {}
for (min_speakers, max_speakers) in [[1, 10], [1, 20]]:
result_for_trainset = {}
for feature_type in self.feature_options:
result_for_feature = {}
# Load best performing model
network = RNN()
name = f'./trained_networks_with_augmentation/rnn_train_{min_speakers}_{max_speakers}/{feature_type}'
network.load_from_file(name)
# Test performance
for test_name, test_data_current in data.items():
x, y = test_data_current['x'], test_data_current['y']
result_for_feature[test_name] = self.__test_net(
network, x, y, feature_type)
result_for_trainset[feature_type] = result_for_feature
result[
f'train_{min_speakers}_{max_speakers}'] = result_for_trainset
with open('experiment_networks_tested.json', 'w+') as fp:
json.dump(result, fp)
return result
