【机器学习】人工神经网络ANN

#一个感知机的例子
import numpy as np
from sklearn.datasets import load_iris
from sklearn.linear_model import Perceptron
iris = load_iris()
X = iris.data[:, (2, 3)] # petal length, petal width
y = (iris.target == 0).astype(np.int) # Iris Setosa?
per_clf = Perceptron(random_state=42)
per_clf.fit(X, y)
y_pred = per_clf.predict([[2, 0.5]]

def heaviside(z):
    return (z >= 0).astype(z.dtype)
def sigmoid(z):
    return 1/(1+np.exp(-z))
#做了多层activation，手工配置权重
def mlp_xor(x1, x2, activation=heaviside):
    return activation(-activation(x1 + x2 - 1.5) + activation(x1 + x2 - 0.5) - 0.5)

import tensorflow as tf

# construction phase
n_inputs = 28*28 # MNIST
# 隐藏层节点数目
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")
def neuron_layer(X, n_neurons, name, activation=None):
    with tf.name_scope(name):
        n_inputs = int(X.get_shape()[1])
        # 标准差初始设定，研究证明设为以下结果训练更快
        stddev = 2 / np.sqrt(n_inputs)
        # 使用截断的正态分布，过滤掉极端的数据，做了一个初始权重矩阵，是input和neurons的全连接矩阵
        init = tf.truncated_normal((n_inputs, n_neurons), stddev=stddev)
        W = tf.Variable(init, name="weights")
        # biases项初始化为0
        b = tf.Variable(tf.zeros([n_neurons]), name="biases")
        # 该层输出
        z = tf.matmul(X, W) + b
        # 根据activation选择激活函数
        if activation=="relu":
            return tf.nn.relu(z)
        else:
            return z

with tf.name_scope("dnn"):
# 算上输入层一共4层的dnn结构
    hidden1 = neuron_layer(X, n_hidden1, "hidden1", activation="relu")
    hidden2 = neuron_layer(hidden1, n_hidden2, "hidden2", activation="relu")
    # 直接输出最后结果值
    logits = neuron_layer(hidden2, n_outputs, "outputs")

# 使用TensorFlow自带函数实现，最新修改成dense函数
from tensorflow.contrib.layers import fully_connected
with tf.name_scope("dnn"):
    hidden1 = fully_connected(X, n_hidden1, scope="hidden1")
    hidden2 = fully_connected(hidden1, n_hidden2, scope="hidden2")
    logits = fully_connected(hidden2, n_outputs, scope="outputs", activation_fn=None)

# 使用logits（网络输出）计算交叉熵，取均值为误差
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()

# Execution Phase
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("/tmp/data/")
# 外层大循环跑400次，每个循环中小循环数据量50
n_epochs = 400
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})
        acc_train = accuracy.eval(feed_dict={X: X_batch, y: y_batch})
        acc_test = accuracy.eval(feed_dict={X: mnist.test.images,y: mnist.test.labels})
         print(epoch, "Train accuracy:", acc_train, "Test accuracy:", acc_test)

# 下次再跑模型时不用再次训练了
save_path = saver.save(sess, "./my_model_final.ckpt")

# 下次调用
with tf.Session() as sess:
    saver.restore(sess, "./my_model_final.ckpt") # or better, use save_path
    X_new_scaled = mnist.test.images[:20]
    Z = logits.eval(feed_dict={X: X_new_scaled})
    y_pred = np.argmax(Z, axis=1)