def setUpClass(self):
self.X, self.y = get_mnist()
self.model = Model(lr=0.01, n_epoch=3, loss=SCCE(),
metrics=['loss', 'accuracy'], optimizer=RMSProp())
self.model.add_layer(Dense(200, inshape=784, activation=ReLU()))
self.model.add_layer(Dense(10, activation=Softmax()))
self.fit_metrics = self.model.fit(self.X, self.y)
class TestMNIST(unittest.TestCase):
@classmethod
def setUpClass(self):
self.X, self.y = get_mnist()
self.model = Model(lr=0.01, n_epoch=3, loss=SCCE(),
metrics=['loss', 'accuracy'], optimizer=RMSProp())
self.model.add_layer(Dense(200, inshape=784, activation=ReLU()))
self.model.add_layer(Dense(10, activation=Softmax()))
self.fit_metrics = self.model.fit(self.X, self.y)
def test_training_accuracy_above_ninety(self):
self.assertGreater(self.fit_metrics['train']['accuracy'], 0.9)
def test_validation_accuracy_above_ninety(self):
self.assertGreater(self.fit_metrics['val']['accuracy'], 0.9)
def setUpClass(self):
self.X, self.y = get_mnist()
np.random.seed(100)
self.X = np.random.permutation(self.X)[:1000]
np.random.seed(100)
self.y = np.random.permutation(self.y)[:1000]
self.model = Model(lr=0.001, n_epoch=100, batch_size=3, loss=SCCE(),
metrics=['loss', 'accuracy'], optimizer=SGD())
self.model.add_layer(Convolution(1, (3, 3), inshape=(None, 1, 28, 28)))
self.model.add_layer(MeanPooling((2, 2)))
self.model.add_layer(Convolution(2, (4, 4)))
self.model.add_layer(MeanPooling((2, 2)))
self.model.add_layer(Flatten())
self.model.add_layer(Dense(10, activation=Softmax()))
self.fit_metrics = self.model.fit(self.X, self.y)
def setUpClass(cls):
cls.model = Model(progress=False,
validation_split=0,
batch_size=4,
metrics=['loss', 'accuracy'],
log_interval=5000)
cls.model.add_layer(Dense(3, inshape=2, activation=Sigmoid()))
cls.model.add_layer(Dense(1, activation=Sigmoid()))
cls.model.compile()
cls.X_train = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
cls.Y_train = np.array([[0], [1], [1], [0]])
cls.model.fit(cls.X_train, cls.Y_train)
class TestMNISTWithConvnet(unittest.TestCase):
@classmethod
def setUpClass(self):
self.X, self.y = get_mnist()
np.random.seed(100)
self.X = np.random.permutation(self.X)[:1000]
np.random.seed(100)
self.y = np.random.permutation(self.y)[:1000]
self.model = Model(lr=0.001, n_epoch=100, batch_size=3, loss=SCCE(),
metrics=['loss', 'accuracy'], optimizer=SGD())
self.model.add_layer(Convolution(1, (3, 3), inshape=(None, 1, 28, 28)))
self.model.add_layer(MeanPooling((2, 2)))
self.model.add_layer(Convolution(2, (4, 4)))
self.model.add_layer(MeanPooling((2, 2)))
self.model.add_layer(Flatten())
self.model.add_layer(Dense(10, activation=Softmax()))
self.fit_metrics = self.model.fit(self.X, self.y)
def test_training_accuracy_above_ninety(self):
self.assertGreater(self.fit_metrics['train']['accuracy'], 0.8)
def test_validation_accuracy_above_ninety(self):
self.assertGreater(self.fit_metrics['val']['accuracy'], 0.8)
from slugnet.activation import ReLU, Softmax
from slugnet.layers import Dense, Dropout
from slugnet.loss import SoftmaxCategoricalCrossEntropy as SCCE
from slugnet.model import Model
from slugnet.optimizers import RMSProp
from slugnet.data.mnist import get_mnist
X, y = get_mnist()
model = Model(lr=0.01,
n_epoch=3,
loss=SCCE(),
metrics=['loss', 'accuracy'],
optimizer=RMSProp())
model.add_layer(Dense(200, inshape=784, activation=ReLU()))
model.add_layer(Dropout(0.5))
model.add_layer(Dense(10, activation=Softmax()))
model.fit(X, y)
from slugnet.activation import ReLU, Softmax
from slugnet.layers import Convolution, Dense, MeanPooling, Flatten
from slugnet.loss import SoftmaxCategoricalCrossEntropy as SCCE
from slugnet.model import Model
from slugnet.optimizers import SGD
from slugnet.data.mnist import get_mnist
X, y = get_mnist()
X = X.reshape((-1, 1, 28, 28)) / 255.0
np.random.seed(100)
X = np.random.permutation(X)[:1000]
np.random.seed(100)
y = np.random.permutation(y)[:1000]
model = Model(lr=0.001,
n_epoch=100,
batch_size=3,
loss=SCCE(),
metrics=['loss', 'accuracy'],
optimizer=SGD())
model.add_layer(Convolution(1, (3, 3), inshape=(None, 1, 28, 28)))
model.add_layer(MeanPooling((2, 2)))
model.add_layer(Convolution(2, (4, 4)))
model.add_layer(MeanPooling((2, 2)))
model.add_layer(Flatten())
model.add_layer(Dense(10, activation=Softmax()))
model.fit(X, y)