v_dim = len(v_xs[0])
som_a = SOM(5,
5,
a_dim,
checkpoint_dir=soma_path,
n_iterations=100,
batch_size=4)
som_v = SOM(5,
5,
v_dim,
checkpoint_dir=somv_path,
n_iterations=100,
batch_size=4)
som_a.train(a_xs)
som_v.train(v_xs)
som_a.memorize_examples_by_class(a_xs, a_ys)
som_v.memorize_examples_by_class(v_xs, v_ys)
hebbian_model = HebbianModel(som_a,
som_v,
a_dim=a_dim,
v_dim=v_dim,
n_presentations=1,
learning_rate=1,
n_classes=n_classes,
checkpoint_dir=hebbian_path)
print('Training...')
hebbian_model.train(a_xs, v_xs)
print('Evaluating...')
accuracy = hebbian_model.evaluate(a_xs,
v_xs,
a_ys,
for n in range(1, 15):
hebbian_model = HebbianModel(som_a,
som_v,
a_dim=a_dim,
v_dim=v_dim,
n_presentations=n,
checkpoint_dir=hebbian_path,
learning_rate=args.lr)
a_xs_fold, v_xs_fold, a_ys_fold, v_ys_fold = create_folds(a_xs_train,
v_xs_train,
a_ys_train,
v_ys_train,
n_folds=n)
# prepare the soms for alternative matching strategies - this is not necessary
# if prediction_alg='regular' in hebbian_model.evaluate(...) below
som_a.memorize_examples_by_class(a_xs_train, a_ys_train)
som_v.memorize_examples_by_class(v_xs_train, v_ys_train)
print('Training...')
hebbian_model.train(a_xs_fold, v_xs_fold)
print('Evaluating...')
accuracy_a = hebbian_model.evaluate(a_xs_test,
v_xs_test,
a_ys_test,
v_ys_test,
source='a',
prediction_alg=args.algo)
accuracy_v = hebbian_model.evaluate(a_xs_test,
v_xs_test,
a_ys_test,
v_ys_test,
source='v',