def part_3_4():
"""
Test hopfield network's robustness on different settings of noisy data
:param p: dictionary of different patterns
"""
# Train with 3 patterns
p_train = [p[0].flatten(), p[1].flatten(), p[2].flatten()]
train_data = np.asarray(p_train)
h_net = HopfieldNet(train_data)
h_net.batch_train()
# Choose a pattern and add noise to it
test_pattern = 2 # Choose between 0, 1, 2
p_test = [p[test_pattern].flatten()]
test_data = np.asarray(p_test)
# Set noise percentages to test on [start, end, step]
noise_percentages = np.arange(0, 101, 1)
n_runs = 1
runs = []
for run in range(n_runs):
acc = {}
# Test for different percentages of noise
for noise_perc in noise_percentages:
# add noise to test data
noisy_test_data = add_noise(test_data[0], noise_perc)
# try to recall
test_pred = h_net.recall([noisy_test_data], epochs=batch_epochs)
acc[noise_perc] = calc_acc(test_data[0], test_pred[0])
if show_plots:
test_pred_1 = test_pred[0].reshape(32,
32) # prepare for plotting
show_tested(noisy_test_data,
test_pred_1,
test_pred_1.shape[0],
test_pred_1.shape[1],
title="Testing with " + str(noise_perc) +
"% noise")
# plot_accuracy(acc)
runs.append(acc)
average_acc = {}
for noise_perc in acc.keys():
av_acc_i = 0.
for run in range(n_runs):
av_acc_i += runs[run][noise_perc]
average_acc[noise_perc] = av_acc_i / float(n_runs)
plot_accuracy(average_acc)
def main():
""" Test for the hopfield network """
"""
This is just for testing the functions
"""
x1 = np.array([1, 1, 1, 1, -1, -1, 1, 1, 1])
x2 = np.array([1, -1, 1, 1, 1, 1, 1, -1, 1])
x3 = np.array([-1, 1, -1, -1, 1, -1, -1, 1, -1])
train_set = np.vstack((x1, x2))
train_set = np.vstack((train_set, x3))
params = {
"epochs": 100,
"neurons": len(x1),
"learn_method": 'classic'
}
hop = hop_net.HopfieldNet(train_set, **params)
hop.batch_train()
show_trained(train_set)
x4d = [1,1,1,1,1,1,1,1,1]
x5d = [1,1,1,1,-1,-1,1,-1,-1]
x45d = np.vstack((x4d, x5d))
test_set = np.vstack((x45d, train_set))
recalled_set = hop.recall(test_set)
for i in range(test_set.shape[0]):
show_tested(test_set[i], recalled_set[i])
def main():
""" main function"""
x1 = np.array([-1, -1, 1, -1, 1, -1, -1, 1])
x2 = np.array([-1, -1, -1, -1, -1, 1, -1, -1])
x3 = np.array([-1, 1, 1, -1, -1, 1, -1, 1])
train_set = np.vstack((x1, x2, x3))
hop = hop_net.HopfieldNet(train_set)
hop.batch_train()
show_trained(train_set, 4, 2)
x1d = np.array([1, -1, 1, -1, 1, -1, -1, 1])
x2d = np.array([1, 1, -1, -1, -1, 1, -1, -1])
x3d = np.array([1, 1, 1, -1, 1, 1, -1, 1])
test_set = np.vstack((train_set, x1d, x2d, x3d))
recalled_set = hop.recall(test_set, epochs=15)
for i in range(test_set.shape[0]):
show_tested(test_set[i], recalled_set[i], 4, 2)
def test_recovery(train_data, method):
"""
Test recovery abilities
:param train_data:
:param method:
:return:
"""
h_net = HopfieldNet(train_data)
h_net.batch_train()
p_test = [p[9].flatten(), p[10].flatten()]
test_data = np.asarray(p_test)
if method == 'batch':
print("Testing recovery with batch")
test_pred = h_net.recall(test_data, epochs=batch_epochs)
else:
print("Testing recovery with sequential")
test_pred, _ = h_net.sequential_recall(test_data,
epochs=seq_epochs,
plot_at_100=plot_at_100)
test_pred_10 = test_pred[0].reshape(32, 32) # prepare for plotting
test_pred_11 = test_pred[1].reshape(32, 32) # prepare for plotting
show_tested(p_test[0],
test_pred_10,
test_pred_10.shape[0],
test_pred_10.shape[1],
title="Testing recovery abilities with " + method)
show_tested(p_test[1],
test_pred_11,
test_pred_11.shape[0],
test_pred_11.shape[1],
title="Testing recovery abilities with " + method)
def test_stability(train_data, method):
"""
Test that the patterns are stable
:param train_data:
:param method:
:return:
"""
h_net = HopfieldNet(train_data)
h_net.batch_train()
test_data = train_data
if method == 'batch':
print("Testing stability with batch")
test_pred = h_net.recall(test_data, epochs=batch_epochs)
else:
print("Testing stability with sequential")
test_pred, _ = h_net.sequential_recall(test_data,
epochs=seq_epochs,
plot_at_100=plot_at_100)
test_pred_1 = test_pred[0].reshape(32, 32) # prepare for plotting
test_pred_2 = test_pred[1].reshape(32, 32) # prepare for plotting
test_pred_3 = test_pred[2].reshape(32, 32) # prepare for plotting
show_tested(test_data[0],
test_pred_1,
test_pred_1.shape[0],
test_pred_1.shape[1],
title="Testing stability with " + method)
show_tested(test_data[1],
test_pred_2,
test_pred_2.shape[0],
test_pred_2.shape[1],
title="Testing stability with " + method)
show_tested(test_data[2],
test_pred_3,
test_pred_3.shape[0],
test_pred_3.shape[1],
title="Testing stability with " + method)