def templateSyncEnsembleAllocation(self, stimulus, steps, connection,
amount_neighbors, analysed_iterations,
expected_length_ensembles):
testing_result = True
for _ in range(3):
network_instance = cnn_network(len(stimulus), connection,
amount_neighbors)
assert len(stimulus) == len(network_instance)
output_dynamic = network_instance.simulate(steps, stimulus)
ensembles = output_dynamic.allocate_sync_ensembles(
analysed_iterations)
obtained_ensemble_sizes = [
len(ensemble) for ensemble in ensembles
]
# critical checks - always determined
assert len(stimulus) == len(network_instance)
assert len(stimulus) == sum(obtained_ensemble_sizes)
if (expected_length_ensembles != None):
obtained_ensemble_sizes.sort()
expected_length_ensembles.sort()
if (obtained_ensemble_sizes != expected_length_ensembles):
continue
assert testing_result == True
def testChaoticNeuralNetwork3DVisualization(self):
stimulus = read_sample(SIMPLE_SAMPLES.SAMPLE_SIMPLE11)
network_instance = cnn_network(len(stimulus))
output_dynamic = network_instance.simulate(10, stimulus)
network_instance.show_network()
cnn_visualizer.show_dynamic_matrix(output_dynamic)
cnn_visualizer.show_observation_matrix(output_dynamic)
cnn_visualizer.show_output_dynamic(output_dynamic)
def testChaoticNeuralNetwork3DVisualization(self):
stimulus = read_sample(SIMPLE_SAMPLES.SAMPLE_SIMPLE11);
network_instance = cnn_network(len(stimulus));
output_dynamic = network_instance.simulate(10, stimulus);
network_instance.show_network();
cnn_visualizer.show_dynamic_matrix(output_dynamic);
cnn_visualizer.show_observation_matrix(output_dynamic);
cnn_visualizer.show_output_dynamic(output_dynamic);
def template_dynamic_cnn(num_osc, steps, stimulus, neighbors, connection, show_network = False):
network_instance = cnn_network(num_osc, connection, amount_neighbors = neighbors);
output_dynamic = network_instance.simulate(steps, stimulus);
print(output_dynamic.allocate_sync_ensembles(10));
if (show_network is True):
network_instance.show_network();
cnn_visualizer.show_output_dynamic(output_dynamic);
cnn_visualizer.show_dynamic_matrix(output_dynamic);
cnn_visualizer.show_observation_matrix(output_dynamic);
def template_dynamic_cnn(num_osc, steps, stimulus, neighbors, connection, show_network = False):
network_instance = cnn_network(num_osc, connection, amount_neighbors = neighbors);
output_dynamic = network_instance.simulate(steps, stimulus);
print(output_dynamic.allocate_sync_ensembles(10));
if (show_network is True):
network_instance.show_network();
cnn_visualizer.show_output_dynamic(output_dynamic);
cnn_visualizer.show_dynamic_matrix(output_dynamic);
cnn_visualizer.show_observation_matrix(output_dynamic);
def templateSyncEnsembleAllocation(self, stimulus, steps, connection, amount_neighbors, analysed_iterations, expected_length_ensembles):
testing_result = True;
for _ in range(3):
network_instance = cnn_network(len(stimulus), connection, amount_neighbors);
assert len(stimulus) == len(network_instance);
output_dynamic = network_instance.simulate(steps, stimulus);
ensembles = output_dynamic.allocate_sync_ensembles(analysed_iterations);
obtained_ensemble_sizes = [len(ensemble) for ensemble in ensembles];
# critical checks - always determined
assert len(stimulus) == len(network_instance);
assert len(stimulus) == sum(obtained_ensemble_sizes);
if (expected_length_ensembles != None):
obtained_ensemble_sizes.sort();
expected_length_ensembles.sort();
if (obtained_ensemble_sizes != expected_length_ensembles):
continue;
assert testing_result == True;
