tensorflow 基础定义

作为TensorFlow的小白，还有很多东西要学的。

(1)    node

node1 = tf.constant(3.0, tf.float32)
node2 = tf.constant(4.0) # also tf.float32 implicitly
print(node1, node2)

 To actuallyevaluate the nodes, we must run the computational graph within a session.

sess = tf.Session()
print(sess.run([node1, node2]))

[3.0,4.0]

 

node3 = tf.add(node1, node2)
print("node3: ", node3)
print("sess.run(node3): ",sess.run(node3))

node3:  Tensor("Add_2:0", shape=(), dtype=float32)
sess.run(node3):  7.0

(2) to accept externalinputs, known as placeholders.

 

a = tf.placeholder(tf.float32)
b = tf.placeholder(tf.float32)
adder_node = a + b  # + provides a shortcut for tf.add(a, b)

tospecify Tensors that provide concrete values to these placeholders:

print(sess.run(adder_node,{a:3, b:4.5}))
print(sess.run(adder_node,{a:[1,3], b:[2,4]}))

输出：7.5
[ 3.  7.]

add_and_triple = adder_node *3.
print(sess.run(add_and_triple,{a:3, b:4.5}))

(3) Variables

W = tf.Variable([.3], tf.float32)
b = tf.Variable([-.3], tf.float32)
x = tf.placeholder(tf.float32)
linear_model = W * x + b

Toinitialize all the variables in a TensorFlow program, you must explicitly calla special operation as follows:

init = tf.global_variables_initializer()
sess.run(init)

print(sess.run(linear_model,{x:[1,2,3,4]}))

[0.          0.30000001  0.60000002  0.90000004]

(4)  loss function 

y = tf.placeholder(tf.float32)
squared_deltas = tf.square(linear_model - y)
loss = tf.reduce_sum(squared_deltas)
print(sess.run(loss,{x:[1,2,3,4], y:[0,-1,-2,-3]}))

(5)    improve w,b

fixW = tf.assign(W,[-1.])
fixb = tf.assign(b, [1.])
sess.run([fixW, fixb])
print(sess.run(loss,{x:[1,2,3,4], y:[0,-1,-2,-3]}))

loss结果为：

0.0

（6）TensorFlow provides optimizers that slowly change each variable inorder to minimize the loss function. The simplest optimizer is gradientdescent.

 

optimizer = tf.train.GradientDescentOptimizer(0.01)
train = optimizer.minimize(loss)

sess.run(init)# reset values to incorrect defaults.
for i in range(1000):
   sess.run(train, {x:[1,2,3,4], y:[0,-1,-2,-3]})

print(sess.run([W, b]))

输出结果：

[array([-0.9999969], dtype=float32), array([0.99999082],
        dtype=float32)]

总结 node,placeholder