taxi-dnn

dnn.py

from keras.models import Sequential, Model
from keras.layers import Dense, Dropout, Activation
from keras import optimizers
from keras import regularizers

model = Sequential()
model.add(Dense(128, activation='relu', input_dim=len(features)))
model.add(Dense(64))
model.add(Dense(1))


model.compile(loss='mse', optimizer='adam')

history = model.fit(x=X_train,
                    y=y_train,
                    batch_size=500,
                    epochs=300,
                    verbose=1,
                    validation_data=(X_test, y_test),
                    shuffle=True)