[轉]博客園
由於深度神經網絡（DNN）層數很多，每次訓練都是逐層由後至前傳遞。傳遞項<1，梯度可能變得非常小趨於0，以此來訓練網絡幾乎不會有什麼變化，即vanishing gradients problem；或者>1梯度非常大，以此修正網絡會不斷震盪，無法形成一個收斂網絡。因而DNN的訓練中可以形成很多tricks。。
1、初始化權重
起初採用正態分布隨機化初始權重，會使得原本單位的variance逐漸變得非常大。例如下圖的sigmoid函數，靠近0點的梯度近似線性很敏感，但到了，即很強烈的輸入產生木訥的輸出。
採用Xavier initialization，根據fan-in（輸入神經元個數）和fan-out（輸出神經元個數）設置權重。
並設計針對不同激活函數的初始化策略，如下圖（左邊是均態分布，右邊正態分布較為常用）
2、激活函數
一般使用ReLU，但是不能有小於0的輸入（dying ReLUs）
a.Leaky RELU
改進方法Leaky ReLU=max(αx,x)，小於0時保留一點微小特徵。
具體應用
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("/tmp/data/")
reset_graph()
n_inputs = 28 * 28 # MNIST
n_hidden1 = 300
n_hidden2 = 100
n_outputs = 10
X = tf.placeholder(tf.float32, shape=(None, n_inputs), name="X")
y = tf.placeholder(tf.int64, shape=(None), name="y")
with tf.name_scope("dnn"):
hidden1 = tf.layers.dense(X, n_hidden1, activation=leaky_relu, name="hidden1")
hidden2 = tf.layers.dense(hidden1, n_hidden2, activation=leaky_relu, name="hidden2")
logits = tf.layers.dense(hidden2, n_outputs, name="outputs")
with tf.name_scope("loss"):
xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=logits)
loss = tf.reduce_mean(xentropy, name="loss")
learning_rate = 0.01
with tf.name_scope("train"):
optimizer = tf.train.GradientDescentOptimizer(learning_rate)
training_op = optimizer.minimize(loss)
with tf.name_scope("eval"):
correct = tf.nn.in_top_k(logits, y, 1)
accuracy = tf.reduce_mean(tf.cast(correct, tf.float32))
init = tf.global_variables_initializer()
saver = tf.train.Saver()
n_epochs = 40
batch_size = 50
with tf.Session() as sess:
init.run()
for epoch in range(n_epochs):
for iteration in range(mnist.train.num_examples // batch_size):
X_batch, y_batch = mnist.train.next_batch(batch_size)
sess.run(training_op, feed_dict={X: X_batch, y: y_batch})
if epoch % 5 == 0:
acc_train = accuracy.eval(feed_dict={X: X_batch, y: y_batch})
acc_test = accuracy.eval(feed_dict={X: mnist.validation.images, y: mnist.validation.labels})
print(epoch, "Batch accuracy:", acc_train, "Validation accuracy:", acc_test)
save_path = saver.save(sess, "./my_model_final.ckpt")
b. ELU改進
另一種改進ELU，在神經元小於0時採用指數變化
#just specify the activation function when building each layer
X = tf.placeholder(tf.float32, shape=(None, n_inputs), name="X")
hidden1 = tf.layers.dense(X, n_hidden1, activation=tf.nn.elu, name="hidden1")
c. SELU
最新提出的是SELU（僅給出關鍵代碼）
with tf.name_scope("dnn"):
hidden1 = tf.layers.dense(X, n_hidden1, activation=selu, name="hidden1")
hidden2 = tf.layers.dense(hidden1, n_hidden2, activation=selu, name="hidden2")
logits = tf.layers.dense(hidden2, n_outputs, name="outputs")
# train 過程
means = mnist.train.images.mean(axis=0, keepdims=True)
stds = mnist.train.images.std(axis=0, keepdims=True) + 1e-10
with tf.Session() as sess:
init.run()
for epoch in range(n_epochs):
for iteration in range(mnist.train.num_examples // batch_size):
X_batch, y_batch = mnist.train.next_batch(batch_size)
X_batch_scaled = (X_batch - means) / stds
sess.run(training_op, feed_dict={X: X_batch_scaled, y: y_batch})
if epoch % 5 == 0:
acc_train = accuracy.eval(feed_dict={X: X_batch_scaled, y: y_batch})
X_val_scaled = (mnist.validation.images - means) / stds
acc_test = accuracy.eval(feed_dict={X: X_val_scaled, y: mnist.validation.labels})
print(epoch, "Batch accuracy:", acc_train, "Validation accuracy:", acc_test)
save_path = saver.save(sess, "./my_model_final_selu.ckpt")
3、Batch Normalization
在2015年，有研究者提出，既然使用mini-batch進行操作，對每一批數據也可採用，在調用激活函數之前，先做一下normalization，使得輸出數據有一個較好的形狀，初始時，超參數scaling（γ）和shifting（β）進行適度縮放平移後傳遞給activation函數。步驟如下：
現今batch normalization已經被TensorFlow實現成一個單獨的層，直接調用
測試時，由於沒有mini-batch，故訓練時直接使用訓練時的mean和standard deviation（），實現代碼如下
import tensorflow as tf
n_inputs = 28 * 28
n_hidden1 = 300
n_hidden2 = 100
n_outputs = 10
batch_norm_momentum = 0.9
X = tf.placeholder(tf.float32, shape=(None, n_inputs), name="X")
y = tf.placeholder(tf.int64, shape=(None), name="y")
training = tf.placeholder_with_default(False, shape=(), name='training')
with tf.name_scope("dnn"):
he_init = tf.contrib.layers.variance_scaling_initializer()
#相當於單獨一層
my_batch_norm_layer = partial(
tf.layers.batch_normalization,
training=training,
momentum=batch_norm_momentum)
my_dense_layer = partial(
tf.layers.dense,
kernel_initializer=he_init)
hidden1 = my_dense_layer(X, n_hidden1, name="hidden1")
bn1 = tf.nn.elu(my_batch_norm_layer(hidden1))# 激活函數使用ELU
hidden2 = my_dense_layer(bn1, n_hidden2, name="hidden2")
bn2 = tf.nn.elu(my_batch_norm_layer(hidden2))
logits_before_bn = my_dense_layer(bn2, n_outputs, name="outputs")
logits = my_batch_norm_layer(logits_before_bn)# 輸出層也做一個batch normalization
with tf.name_scope("loss"):
xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=logits)
loss = tf.reduce_mean(xentropy, name="loss")
with tf.name_scope("train"):
optimizer = tf.train.GradientDescentOptimizer(learning_rate)
training_op = optimizer.minimize(loss)
with tf.name_scope("eval"):
correct = tf.nn.in_top_k(logits, y, 1)
accuracy = tf.reduce_mean(tf.cast(correct, tf.float32))
init = tf.global_variables_initializer()
saver = tf.train.Saver()
n_epochs = 20
batch_size = 200
#需要顯示調用訓練時得出的方差均值，需要額外調用這些運算元
extra_update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
#在training和testing時不一樣
with tf.Session() as sess:
init.run()
for epoch in range(n_epochs):
for iteration in range(mnist.train.num_examples // batch_size):
X_batch, y_batch = mnist.train.next_batch(batch_size)
sess.run([training_op, extra_update_ops],
feed_dict={training: True, X: X_batch, y: y_batch})
accuracy_val = accuracy.eval(feed_dict={X: mnist.test.images,
y: mnist.test.labels})
print(epoch, "Test accuracy:", accuracy_val)
save_path = saver.save(sess, "./my_model_final.ckpt")
4、Gradient Clipp
處理gradient之後往後傳，一定程度上解決梯度爆炸問題。（但由於有了batch normalization，此方法用的不多）
threshold = 1.0
optimizer = tf.train.GradientDescentOptimizer(learning_rate)
grads_and_vars = optimizer.compute_gradients(loss)
capped_gvs = [(tf.clip_by_value(grad, -threshold, threshold), var)
for grad, var in grads_and_vars]
training_op = optimizer.apply_gradients(capped_gvs)
5、重用之前訓練過的層（Reusing Pretrained Layers）
對之前訓練的模型稍加修改，節省時間，在深度模型訓練（由於有很多層）中經常使用。
一般相似問題，分類數等和問題緊密相關的output層與最後一個直接與output相關的隱層不可以直接用，仍需自己訓練。
如下圖所示，在已訓練出一個複雜net後，遷移到相對簡單的net時，hidden1和2固定不動，hidden3稍作變化，hidden4和output自己訓練。。這在沒有自己GPU情況下是非常節省時間的做法。
# 只選取需要的操作
X = tf.get_default_graph().get_tensor_by_name("X:0")
y = tf.get_default_graph().get_tensor_by_name("y:0")
accuracy = tf.get_default_graph().get_tensor_by_name("eval/accuracy:0")
training_op = tf.get_default_graph().get_operation_by_name("GradientDescent")
# 如果你是原模型的作者，可以賦給模型一個清楚的名字保存下來
for op in (X, y, accuracy, training_op):
tf.add_to_collection("my_important_ops", op)
# 如果你要使用這個模型
X, y, accuracy, training_op = tf.get_collection("my_important_ops")
# 訓練時
with tf.Session() as sess:
saver.restore(sess, "./my_model_final.ckpt")
for epoch in range(n_epochs):
for iteration in range(mnist.train.num_examples // batch_size):
X_batch, y_batch = mnist.train.next_batch(batch_size)
sess.run(training_op, feed_dict={X: X_batch, y: y_batch})
accuracy_val = accuracy.eval(feed_dict={X: mnist.test.images,
y: mnist.test.labels})
print(epoch, "Test accuracy:", accuracy_val)
save_path = saver.save(sess, "./my_new_model_final.ckpt")
a. Freezing the Lower Layers
訓練時固定底層參數，達到Freezing the Lower Layers的目的
# 以MINIST為例
n_inputs = 28 * 28 # MNIST
n_hidden1 = 300 # reused
n_hidden2 = 50 # reused
n_hidden3 = 50 # reused
n_hidden4 = 20 # new!
n_outputs = 10 # new!
X = tf.placeholder(tf.float32, shape=(None, n_inputs), name="X")
y = tf.placeholder(tf.int64, shape=(None), name="y")
with tf.name_scope("dnn"):
hidden1 = tf.layers.dense(X, n_hidden1, activation=tf.nn.relu,
name="hidden1") # reused frozen
hidden2 = tf.layers.dense(hidden1, n_hidden2, activation=tf.nn.relu,
name="hidden2") # reused frozen
hidden2_stop = tf.stop_gradient(hidden2)
hidden3 = tf.layers.dense(hidden2_stop, n_hidden3, activation=tf.nn.relu,
name="hidden3") # reused, not frozen
hidden4 = tf.layers.dense(hidden3, n_hidden4, activation=tf.nn.relu,
name="hidden4") # new!
logits = tf.layers.dense(hidden4, n_outputs, name="outputs") # new!
with tf.name_scope("loss"):
xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=logits)
loss = tf.reduce_mean(xentropy, name="loss")
with tf.name_scope("eval"):
correct = tf.nn.in_top_k(logits, y, 1)
accuracy = tf.reduce_mean(tf.cast(correct, tf.float32), name="accuracy")
with tf.name_scope("train"):
optimizer = tf.train.GradientDescentOptimizer(learning_rate)
training_op = optimizer.minimize(loss)
reuse_vars = tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES,
scope="hidden[123]") # regular expression
reuse_vars_dict = dict([(var.op.name, var) for var in reuse_vars])
restore_saver = tf.train.Saver(reuse_vars_dict) # to restore layers 1-3
init = tf.global_variables_initializer()
saver = tf.train.Saver()
with tf.Session() as sess:
init.run()
restore_saver.restore(sess, "./my_model_final.ckpt")
for epoch in range(n_epochs):
for iteration in range(mnist.train.num_examples // batch_size):
X_batch, y_batch = mnist.train.next_batch(batch_size)
sess.run(training_op, feed_dict={X: X_batch, y: y_batch})
accuracy_val = accuracy.eval(feed_dict={X: mnist.test.images,
y: mnist.test.labels})
print(epoch, "Test accuracy:", accuracy_val)
save_path = saver.save(sess, "./my_new_model_final.ckpt")
b. Catching the Frozen Layers
訓練時直接從lock層之後的層開始訓練，Catching the Frozen Layers
# 以MINIST為例
n_inputs = 28 * 28 # MNIST
n_hidden1 = 300 # reused
n_hidden2 = 50 # reused
n_hidden3 = 50 # reused
n_hidden4 = 20 # new!
n_outputs = 10 # new!
X = tf.placeholder(tf.float32, shape=(None, n_inputs), name="X")
y = tf.placeholder(tf.int64, shape=(None), name="y")
with tf.name_scope("dnn"):
hidden1 = tf.layers.dense(X, n_hidden1, activation=tf.nn.relu,
name="hidden1") # reused frozen
hidden2 = tf.layers.dense(hidden1, n_hidden2, activation=tf.nn.relu,
name="hidden2") # reused frozen & cached
hidden2_stop = tf.stop_gradient(hidden2)
hidden3 = tf.layers.dense(hidden2_stop, n_hidden3, activation=tf.nn.relu,
name="hidden3") # reused, not frozen
hidden4 = tf.layers.dense(hidden3, n_hidden4, activation=tf.nn.relu,
name="hidden4") # new!
logits = tf.layers.dense(hidden4, n_outputs, name="outputs") # new!
with tf.name_scope("loss"):
xentropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=logits)
loss = tf.reduce_mean(xentropy, name="loss")
with tf.name_scope("eval"):
correct = tf.nn.in_top_k(logits, y, 1)
accuracy = tf.reduce_mean(tf.cast(correct, tf.float32), name="accuracy")
with tf.name_scope("train"):
optimizer = tf.train.GradientDescentOptimizer(learning_rate)
training_op = optimizer.minimize(loss)
reuse_vars = tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES,
scope="hidden[123]") # regular expression
reuse_vars_dict = dict([(var.op.name, var) for var in reuse_vars])
restore_saver = tf.train.Saver(reuse_vars_dict) # to restore layers 1-3
init = tf.global_variables_initializer()
saver = tf.train.Saver()
import numpy as np
n_batches = mnist.train.num_examples // batch_size
with tf.Session() as sess:
init.run()
restore_saver.restore(sess, "./my_model_final.ckpt")
h2_cache = sess.run(hidden2, feed_dict={X: mnist.train.images})
h2_cache_test = sess.run(hidden2, feed_dict={X: mnist.test.images}) # not shown in the book
for epoch in range(n_epochs):
shuffled_idx = np.random.permutation(mnist.train.num_examples)
hidden2_batches = np.array_split(h2_cache[shuffled_idx], n_batches)
y_batches = np.array_split(mnist.train.labels[shuffled_idx], n_batches)
for hidden2_batch, y_batch in zip(hidden2_batches, y_batches):
sess.run(training_op, feed_dict={hidden2:hidden2_batch, y:y_batch})
accuracy_val = accuracy.eval(feed_dict={hidden2: h2_cache_test, # not shown
y: mnist.test.labels}) # not shown
print(epoch, "Test accuracy:", accuracy_val) # not shown
save_path = saver.save(sess, "./my_new_model_final.ckpt")
6、Unsupervised Pretraining
該方法的提出，讓人們對深度學習網絡的訓練有了一個新的認識，可以利用不那麼昂貴的未標註數據，訓練數據時沒有標註的數據先做一個Pretraining訓練出一個差不多的網絡，再使用帶label的數據做正式的訓練進行反向傳遞，增進深度模型可用性
也可以在相似模型中做pretraining
7、Faster Optimizers
在傳統的SGD上提出改進
有Momentum optimization（最早提出，利用慣性衝量），Nesterov Accelerated Gradient，AdaGrad（adaptive gradient每層下降不一樣），RMSProp，Adam optimization（結合adagrad和momentum，用的最多，是預設的optimizer）
a. momentum optimization
記住之前算出的gradient方向，作為慣性加到當前梯度上。相當於下山時，SGD是靜止的之判斷當前最陡的是哪裡，而momentum相當於在跑的過程中不斷修正方向，顯然更加有效。
b. Nesterov Accelerated Gradient
只計算當前這點的梯度，超前一步，再往前跑一點計算會更准一些。
c. AdaGrad
各個維度計算梯度作為分母，加到當前梯度上，不同維度梯度下降不同。如下圖所示，橫軸比縱軸平緩很多，傳統gradient僅僅單純沿法線方向移動，而AdaGrad平緩的θ1走的慢點，陡的θ2走的快點，效果較好。
但也有一定缺陷，s不斷積累，分母越來越大，可能導致最後走不動。
d. RMSProp（Adadelta）
只加一部分，加一個衰減係數只選取相關的最近幾步相關係數
e. Adam Optimization
目前用的最多效果最好的方法，結合AdaGrad和Momentum的優點
# TensorFlow中調用方法
optimizer = tf.train.MomentumOptimizer(learning_rate=learning_rate,momentum=0.9)
optimizer = tf.train.MomentumOptimizer(learning_rate=learning_rate,momentum=0.9, use_nesterov=True)
optimizer = tf.train.RMSPropOptimizer(learning_rate=learning_rate,momentum=0.9, decay=0.9, epsilon=1e-10)
# 可以看出AdamOptimizer最省心了
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)
8、learning rate scheduling
learning rate的設置也很重要，如下圖所示，太大不會收斂到全局最優，太小收斂效果最差。最理想情況是都一定情況縮小learning rate，先大後小
a. Exponential Scheduling
指數級下降學習率
initial_learning_rate = 0.1
decay_steps = 10000
decay_rate = 1/10
global_step = tf.Variable(0, trainable=False)
learning_rate = tf.train.exponential_decay(initial_learning_rate, global_step, decay_steps, decay_rate)
optimizer = tf.train.MomentumOptimizer(learning_rate, momentum=0.9)
training_op = optimizer.minimize(loss, global_step=global_step)
9、Avoiding Overfitting Through Regularization
解決深度模型過擬合問題
a. Early Stopping
訓練集上錯誤率開始上升時停止
b. l1和l2正則化
# construct the neural network
base_loss = tf.reduce_mean(xentropy, name="avg_xentropy")
reg_losses = tf.reduce_sum(tf.abs(weights1)) + tf.reduce_sum(tf.abs(weights2))
loss = tf.add(base_loss, scale * reg_losses, name="loss")
with arg_scope( [fully_connected], weights_regularizer=tf.contrib.layers.l1_regularizer(scale=0.01)):
hidden1 = fully_connected(X, n_hidden1, scope="hidden1")
hidden2 = fully_connected(hidden1, n_hidden2, scope="hidden2")
logits = fully_connected(hidden2, n_outputs, activation_fn=None,scope="out")
reg_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
loss = tf.add_n([base_loss] + reg_losses, name="loss")
c. dropout
一種新的正則化方法，隨機生成一個機率，大於某個閾值就扔掉，隨機扔掉一些神經元節點，結果表明dropout很能解決過擬合問題。可強迫現有神經元不會集中太多特徵，降低網絡複雜度，魯棒性增強。
加入dropout後，training和test的準確率會很接近，一定程度解決overfit問題
training = tf.placeholder_with_default(False, shape=(), name='training')
dropout_rate = 0.5 # == 1 - keep_prob
X_drop = tf.layers.dropout(X, dropout_rate, training=training)
with tf.name_scope("dnn"):
hidden1 = tf.layers.dense(X_drop, n_hidden1, activation=tf.nn.relu,
name="hidden1")
hidden1_drop = tf.layers.dropout(hidden1, dropout_rate, training=training)
hidden2 = tf.layers.dense(hidden1_drop, n_hidden2, activation=tf.nn.relu,
name="hidden2")
hidden2_drop = tf.layers.dropout(hidden2, dropout_rate, training=training)
logits = tf.layers.dense(hidden2_drop, n_outputs, name="outputs")
d. Max-Norm Regularization
可以把超出threshold的權重截取掉，一定程度上讓網絡更加穩定
View Code
e. Date Augmentation
深度學習網絡是一個數據饑渴模型，需要很多的數據。擴大數據集，例如圖片左右鏡像翻轉，隨機截取，傾斜隨機角度，變換敏感度，改變色調等方法，擴大數據量，減少overfit可能性
10、default DNN configuration