def test_prob_1d(self):
mlp = MultilayerPerceptron(
num_inputs=4, num_hidden_layers=1, num_hidden_nodes=3)
mlp.fit(X_TRAIN, LABELS_TRAIN, epochnum=5)
pred_prob = mlp.predict_prob(X_TRAIN)
assert np.all(np.logical_and(0 <= pred_prob, pred_prob <= 1))
def test_same_state(self):
self.mlp0 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp1 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_same_state(self):
self.mlp0 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp1 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_prob_1d(self):
mlp = MultilayerPerceptron(num_inputs=4,
num_hidden_layers=1,
num_hidden_nodes=3)
mlp.fit(X_TRAIN, LABELS_TRAIN, epochnum=5)
pred_prob = mlp.predict_prob(X_TRAIN)
assert np.all(np.logical_and(0 <= pred_prob, pred_prob <= 1))
def test_seed_persistence(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=6)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
np.random.normal(size=50)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_seed_persistence(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=6)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
np.random.normal(size=50)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_copy(self):
mlp0 = MultilayerPerceptron()
mlp1 = mlp0.copy()
assert mlp0.params == mlp1.params
check_mlp_same_est(mlp0, mlp1)
for l0, l1 in zip(mlp0.layers, mlp1.layers):
assert not np.may_share_memory(l0.weights, l1.weights)
assert not np.may_share_memory(l0.activations, l1.activations)
assert l0.bias == l1.bias
def test_copy(self):
mlp0 = MultilayerPerceptron()
mlp1 = mlp0.copy()
assert mlp0.params == mlp1.params
check_mlp_same_est(mlp0, mlp1)
for l0, l1 in zip(mlp0.layers, mlp1.layers):
assert not np.may_share_memory(l0.weights, l1.weights)
assert not np.may_share_memory(l0.activations, l1.activations)
assert l0.bias == l1.bias
def test_prob_2d(self):
x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0, 1], [1, 0], [0, 1], [1, 0]])
mlp = MultilayerPerceptron(
num_inputs=3, num_outputs=2,
num_hidden_layers=1)
mlp.fit(x, y, epochnum=5)
pred_prob = mlp.predict_prob(x)
assert np.all(np.logical_and(0 <= pred_prob, pred_prob <= 1))
assert np.allclose(np.sum(pred_prob, axis=1), 1)
def test_easy_multidim_y(self):
x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0, 1], [1, 0], [0, 1], [1, 0]])
mlp = MultilayerPerceptron(
num_inputs=3, num_outputs=2,
num_hidden_layers=1, num_hidden_nodes=6, seed=323440,
learn_rate = .8, learn_rate_evol='constant', momentum=.1
)
mlp.fit(x, y, epochnum=50)
results = mlp.classify(x, max_ind=True)
assert np.allclose(to_dummies(results), y)
def test_prob_2d(self):
x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0, 1], [1, 0], [0, 1], [1, 0]])
mlp = MultilayerPerceptron(num_inputs=3,
num_outputs=2,
num_hidden_layers=1)
mlp.fit(x, y, epochnum=5)
pred_prob = mlp.predict_prob(x)
assert np.all(np.logical_and(0 <= pred_prob, pred_prob <= 1))
assert np.allclose(np.sum(pred_prob, axis=1), 1)
def test_easy(self):
x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([0, 1, 0, 1])
mlp = MultilayerPerceptron(
num_inputs=3, num_hidden_layers=1, num_hidden_nodes=6, seed=323440,
learn_rate = .8, learn_rate_evol='constant', momentum=.1
)
mlp.fit(x, y, add_constant=True, epochnum=100, verbose=True)
print(mlp.classify(x, add_constant=True))
assert np.allclose(mlp.classify(x, add_constant=True).reshape(len(y)),
y)
def test_easy_multidim_y(self):
x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0, 1], [1, 0], [0, 1], [1, 0]])
mlp = MultilayerPerceptron(num_inputs=3,
num_outputs=2,
num_hidden_layers=1,
num_hidden_nodes=6,
seed=323440,
learn_rate=.8,
learn_rate_evol='constant',
momentum=.1)
mlp.fit(x, y, epochnum=50)
results = mlp.classify(x, max_ind=True)
assert np.allclose(to_dummies(results), y)
def test_fit(self):
x = np.column_stack([np.ones(len(X_TRAIN)), X_TRAIN])
mlp = MultilayerPerceptron(num_inputs=4,
num_hidden_layers=1,
num_hidden_nodes=3)
mlp.fit(x, LABELS_TRAIN, epochnum=5, add_constant=False)
mlp.classify(x, add_constant=False)
def test_params(self):
param_dict = {
'num_inputs': 4,
'num_outputs': 2,
'num_hidden_layers': 2,
'num_hidden_nodes': [2, 3],
'learn_rate': .4,
'momentum': .2,
'seed': 24545,
'sigmoid': metrics.tanh,
'weight_init_range': .2
}
mlp = MultilayerPerceptron(**param_dict)
for key, val in param_dict.items():
assert mlp.params[key] == val
def test_easy(self):
x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([0, 1, 0, 1])
mlp = MultilayerPerceptron(num_inputs=3,
num_hidden_layers=1,
num_hidden_nodes=6,
seed=323440,
learn_rate=.8,
learn_rate_evol='constant',
momentum=.1)
mlp.fit(x, y, add_constant=True, epochnum=100, verbose=True)
print(mlp.classify(x, add_constant=True))
assert np.allclose(
mlp.classify(x, add_constant=True).reshape(len(y)), y)
from simpleml.neural import MultilayerPerceptron
from simpleml.transform import to_dummies
# Load data
with gzip.open('../data/mnist.gz', 'rb') as f:
data = pickle.load(f)
data = {key: (val[0], val[1], to_dummies(val[1])) for key, val in data.items()}
# Setup estimator
num_hidden_nodes = [101]
mlp = MultilayerPerceptron(
num_inputs=data['train'][0].shape[1]+1,
num_outputs=data['train'][2].shape[1],
num_hidden_layers=len(num_hidden_nodes), num_hidden_nodes=num_hidden_nodes,
learn_rate=.5, momentum=.1, seed=23456
)
# Estimate multilayer perceptron
start = time.perf_counter()
mlp.fit(data['train'][0], data['train'][2],
epochnum=10, verbose=1)
pred = mlp.classify(data['test'][0], max_ind=True)
print("Time: {:5.2f}, Error: {:5.4f}".format(
time.perf_counter() - start,
1 - np.mean(pred == data['test'][1])
))
# Visualize first hidden layer
# x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
# y = np.array([0, 1, 1, 0])
# Setup mesh for graphing boundary
x0_min, x1_min = np.min(x, axis=0) - .5
x0_max, x1_max = np.max(x, axis=0) + .5
x0, x1 = np.meshgrid(np.linspace(x0_min, x0_max, 500),
np.linspace(x1_min, x1_max, 500))
x_flatmesh = np.column_stack([x0.ravel(), x1.ravel()])
# Setup estimator
mlp = MultilayerPerceptron(num_inputs=3,
num_outputs=1,
num_hidden_layers=1,
num_hidden_nodes=6,
learn_rate=1,
learn_rate_evol='sqrt',
momentum=.1,
seed=23456)
# Estimate and plot
start = time.perf_counter()
with PdfPages('ex_mlp_2d.pdf') as pdf:
for i in range(0, 25):
mlp.fit(x, y, epochnum=20, add_constant=True)
Z = mlp.classify(x_flatmesh, add_constant=True)
Z = Z.reshape(x0.shape)
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
from simpleml.neural import MultilayerPerceptron
from simpleml.transform import to_dummies
# Load data
with gzip.open('../data/mnist.gz', 'rb') as f:
data = pickle.load(f)
data = {key: (val[0], val[1], to_dummies(val[1])) for key, val in data.items()}
# Setup estimator
num_hidden_nodes = [101]
mlp = MultilayerPerceptron(num_inputs=data['train'][0].shape[1] + 1,
num_outputs=data['train'][2].shape[1],
num_hidden_layers=len(num_hidden_nodes),
num_hidden_nodes=num_hidden_nodes,
learn_rate=.5,
momentum=.1,
seed=23456)
# Estimate multilayer perceptron
start = time.perf_counter()
mlp.fit(data['train'][0], data['train'][2], epochnum=10, verbose=1)
pred = mlp.classify(data['test'][0], max_ind=True)
print("Time: {:5.2f}, Error: {:5.4f}".format(
time.perf_counter() - start, 1 - np.mean(pred == data['test'][1])))
# Visualize first hidden layer
fig1 = plt.figure(figsize=(10, 10))
for i in range(num_hidden_nodes[0] - 1):
side = np.sqrt(num_hidden_nodes[0] - 1)
# # XOR
# x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
# y = np.array([0, 1, 1, 0])
# Setup mesh for graphing boundary
x0_min, x1_min = np.min(x, axis=0)-.5
x0_max, x1_max = np.max(x, axis=0)+.5
x0, x1 = np.meshgrid(np.linspace(x0_min, x0_max, 500),
np.linspace(x1_min, x1_max, 500))
x_flatmesh = np.column_stack([x0.ravel(), x1.ravel()])
# Setup estimator
mlp = MultilayerPerceptron(
num_inputs=3, num_outputs=1, num_hidden_layers=1, num_hidden_nodes=6,
learn_rate=1, learn_rate_evol='sqrt', momentum=.1, seed=23456
)
# Estimate and plot
start = time.perf_counter()
with PdfPages('ex_mlp_2d.pdf') as pdf:
for i in range(0, 25):
mlp.fit(x, y, epochnum=20, add_constant=True)
Z = mlp.classify(x_flatmesh, add_constant=True)
Z = Z.reshape(x0.shape)
fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
ax.contour(x0, x1, Z, cmap=plt.cm.Paired)
ax.scatter(x[:,0], x[:,1], c=y, cmap=plt.cm.Paired)
def test_same_seed(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
class TestMLPSeed:
params = {'num_inputs': 4, 'num_hidden_layers': 1, 'num_hidden_nodes': 3}
def test_same_seed(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_same_state(self):
self.mlp0 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp1 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_seed_persistence(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=6)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
np.random.normal(size=50)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_values(case):
mlp = MultilayerPerceptron(learn_rate_evol=case[0])
for arg, out in case[1]:
assert np.isclose(mlp.learn_rate_evol(arg), out)
def test_values(case):
mlp = MultilayerPerceptron(learn_rate_evol=case[0])
for arg, out in case[1]:
assert np.isclose(mlp.learn_rate_evol(arg), out)
def test_nodes(self):
mlp = MultilayerPerceptron(num_hidden_layers=3, num_hidden_nodes=3)
assert mlp.params['num_hidden_nodes'] == [3, 3, 3]
def test_not_recognized(self):
MultilayerPerceptron(learn_rate_evol='unrecognized')
def test_same_seed(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
class TestMLPSeed:
params = {'num_inputs':4, 'num_hidden_layers':1, 'num_hidden_nodes':3}
def test_same_seed(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_same_state(self):
self.mlp0 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp1 = MultilayerPerceptron(seed=np.random.RandomState(345),
**self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_seed_persistence(self):
self.mlp0 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp1 = MultilayerPerceptron(seed=2345, **self.params)
self.mlp0.fit(X_TRAIN, LABELS_TRAIN, epochnum=6)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
np.random.normal(size=50)
self.mlp1.fit(X_TRAIN, LABELS_TRAIN, epochnum=3)
check_mlp_same_est(self.mlp0, self.mlp1)
def test_nodes_wrong_number(self):
MultilayerPerceptron(num_hidden_layers=1, num_hidden_nodes=[3, 3])
def test_print(self):
mlp = MultilayerPerceptron()
print(mlp)
def test_fit(self):
x = np.column_stack([np.ones(len(X_TRAIN)), X_TRAIN])
mlp = MultilayerPerceptron(
num_inputs=4, num_hidden_layers=1, num_hidden_nodes=3)
mlp.fit(x, LABELS_TRAIN, epochnum=5, add_constant=False)
mlp.classify(x, add_constant=False)
