tensorflow
Main classes
- tf.Graph()
- tf.Operation()
- tf.Tensor()
- tf.Session()
Some useful functions
- tf.get_default_session()
- tf.get_default_graph()
- tf.reset_default_graph()
- ops.reset_default_graph()
- tf.device(“/cpu:0”)
- tf.name_scope(value)
- tf.convert_to_tensor(value)
TensorFlow Optimizers
- GradientDescentOptimizer
- AdadeltaOptimizer
- AdagradOptimizer
- MomentumOptimizer
- AdamOptimizer
- FtrlOptimizer
- RMSPropOptimizer
Reduction
- reduce_sum
- reduce_prod
- reduce_min
- reduce_max
- reduce_mean
- reduce_all
- reduce_any
- accumulate_n
Activation functions
- tf.nn
- relu
- relu6
- elu
- softplus
- softsign
- dropout
- bias_add
- sigmoid
- tanh
- sigmoid_cross_entropy_with_logits
- softmax
- log_softmax
- softmax_cross_entropy_with_logits
- sparse_softmax_cross_entropy_with_logits
- weighted_cross_entropy_with_logits
keras
基本例子
import numpy as np
from keras.models import Sequential
from keras.layers import Dense
data = np.random.random((1000,100))
labels = np.random.randint(2,size=(1000,1))
model = Sequential()
model.add(Dense(32,activation='relu',input_dim=100))
model.add(Dense(1, activation='sigmoid'))
model.compile(optimizer='rmsprop',loss='binary_crossentroy', metrics=['accuracy'])
keras数据集
from keras.datasets import boston_housing, mnist, cifar10,imdb
(x_train,y_train),(x_test,y_test) = mnist.load_data()
(x_train2,y_train2),(x_test2,y_test2) = boston_housing.load_data()
(x_train3,y_train3),(x_test3,y_test3) = cifar10.load_data()
(x_train4,y_train4),(x_test4,y_test4) = imdb.load_data(num_words=20000)
num_classes = 10
model.fit(data,labels,epochs=10,batch_size=32)
predictions = model.predict(data)
from urllib.request import urlopen
data = np.loadtxt(urlopen(
"http://archive.ics.uci.edu/ml/machine-learning-databases/pima-indians-diabetes/pima-indians-diabetes.data"),
delimiter=",")
X = data[:,0:8]
y = data [:,8]
模型-顺序模型
from keras.models import Sequential
model = Sequential()
model2 = Sequential()
model3 = Sequential()
模型-多层感知机
#Binary Classification
from keras.layers import Dense
model.add(Dense(12,input_dim=8,kernel_initializer='uniform',activation='relu'))
model.add(Dense(8,kernel_initializer='uniform',activation='relu'))
model.add(Dense(1,kernel_initializer='uniform',activation='sigmoid'))
#Multi-Class Classification
from keras.layers import Dropout
model.add(Dense(512,activation='relu',input_shape=(784,)))
model.add(Dropout(0.2))
model.add(Dense(512,activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10,activation='softmax'))
#Regression
model.add(Dense(64,activation='relu',input_dim=train_data.shape[1]))
model.add(Dense(1))
模型-卷积神经网络
from keras.layers import Activation,Conv2D,MaxPooling2D,Flatten
model2.add(Conv2D(32,(3,3),padding='same',input_shape=x_train.shape[1:]))
model2.add(Activation('relu'))
model2.add(Conv2D(32,(3,3)))
model2.add(Activation('relu'))
model2.add(MaxPooling2D(pool_size=(2,2)))
model2.add(Dropout(0.25))
model2.add(Conv2D(64,(3,3), padding='same'))
model2.add(Activation('relu'))
model2.add(Conv2D(64,(3, 3)))
model2.add(Activation('relu'))
model2.add(MaxPooling2D(pool_size=(2,2)))
model2.add(Dropout(0.25))
model2.add(Flatten())
model2.add(Dense(512))
model2.add(Activation('relu'))
model2.add(Dropout(0.5))
model2.add(Dense(num_classes))
model2.add(Activation('softmax'))
模型-循环神经网络
from keras.klayers import Embedding,LSTM
model3.add(Embedding(20000,128))
model3.add(LSTM(128,dropout=0.2,recurrent_dropout=0.2))
model3.add(Dense(1,activation='sigmoid'))
模型调优
from keras.optimizers import RMSprop
opt = RMSprop(lr=0.0001, decay=1e-6)
model2.compile(loss='categorical_crossentropy',optimizer=opt,metrics=['accuracy'])
early stopping
from keras.callbacks import EarlyStopping
early_stopping_monitor = EarlyStopping(patience=
model3.fit(x_train4, y_train4, batch_size=32, epochs=15,
validation_data=(x_test4,y_test4),callbacks=[early_stopping_monitor])
编译模型
#MLP: Binary Classification
model.compile(optimizer='adam', loss='binary_crossentropy',metrics=['accuracy'])
#MLP: Multi-Class Classification
model.compile(optimizer='rmsprop',loss='categorical_crossentropy',metrics=['accuracy'])
#MLP: Regression
model.compile(optimizer='rmsprop',loss='mse',metrics=['mae'])
#Recurrent Neural Network
model3.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])
保存和加载模型
from keras.models import load_model
model3.save('model_file.h5')
my_model = load_model('my_model.h5')
查看模型详情
model.output_shape #Model output shape
model.summary() #Model summary representation
model.get_config() #Model configuration
model.get_weights() #List all weight tensors in the model
预测
model3.predict(x_test4, batch_size=32)
model3.predict_classes(x_test4,batch_size=32)
训练模型
model3.fit(x_train4,y_train4,batch_size=32,epochs=15,verbose=1,
validation_data=(x_test4,y_test4))
评价模型性能
score = model3.evaluate(x_test,y_test,batch_size=32)
数据预处理
#Sequence Padding
from keras.preprocessing import sequence
x_train4 = sequence.pad_sequences(x_train4,maxlen=80)
x_test4 = sequence.pad_sequences(x_test4,maxlen=80)
#One-Hot Encoding
from keras.utils import to_categorical
Y_train = to_categorical(y_train, num_classes)
Y_test = to_categorical(y_test, num_classes)
Y_train3 = to_categorical(y_train3, num_classes)
Y_test3 = to_categorical(y_test3, num_classes)
#Train and Test Sets
from sklearn.model_selection import train_test_split
X_train5,X_test5,y_train5,y_test5 = train_test_split(X,y,test_size=0.33,
random_state=42)
#Standardization/Normalization
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler().fit(x_train2)
standardized_X = scaler.transform(x_train2)
standardized_X_test = scaler.transform(x_test2)