python
from pydeep.neural_network import NeuralNetwork
nn = NeuralNetwork()
nn.add_layer(10, activation='sigmoid')
nn.add_layer(5, activation='sigmoid')
nn.add_layer(2, activation='softmax')
X_train, y_train = get_training_data()
nn.set_optimizer('adam')
nn.fit(X_train, y_train, epochs=100, batch_size=32)
python
from pydeep.neural_network import CNN
from pydeep.utils import load_image_dataset
X_train, y_train = load_image_dataset('images/train')
X_test, y_test = load_image_dataset('images/test')
cnn = CNN()
cnn.add_convolutional_layer(num_filters=32, filter_size=(3, 3))
cnn.add_max_pooling_layer(pool_size=(2, 2))
cnn.add_dense_layer(units=128, activation='relu')
cnn.add_output_layer(num_classes=10, activation='softmax')
cnn.set_optimizer('adam')
cnn.fit(X_train, y_train, epochs=10, batch_size=32)
score = cnn.evaluate(X_test, y_test)
python
from pydeep.neural_network import RNN
from pydeep.text import load_text_dataset
X_train, y_train = load_text_dataset('reviews/train')
X_test, y_test = load_text_dataset('reviews/test')
rnn = RNN()
rnn.add_embedding_layer(input_length=100, embedding_size=50)
rnn.add_lstm_layer(units=64, return_sequences=True)
rnn.add_lstm_layer(units=32)
rnn.add_output_layer(num_classes=2, activation='softmax')
rnn.set_optimizer('sgd')
rnn.fit(X_train, y_train, epochs=10, batch_size=32)
score = rnn.evaluate(X_test, y_test)
