1 TensorFlow 101

import tensorflow as tf
import numpy as np

Get an Interactive TensorFlow Session

tfs = tf.InteractiveSession()

Customary Hello TensorFlow !!!

hello = tf.constant("Hello TensorFlow !!")
print(tfs.run(hello))

b'Hello TensorFlow !!'

Constants

c1 = tf.constant(5, name='x')
c2 = tf.constant(6.0, name='y')
c3 = tf.constant(7.0, tf.float32, name='z')
print('c1 (x): ', c1)
print('c2 (y): ', c2)
print('c3 (z): ', c3)

c1 (x):  Tensor("x:0", shape=(), dtype=int32)
c2 (y):  Tensor("y:0", shape=(), dtype=float32)
c3 (z):  Tensor("z:0", shape=(), dtype=float32)

print('run([c1,c2,c3]) : ', tfs.run([c1, c2, c3]))

run([c1,c2,c3]) :  [5, 6.0, 7.0]

Operations

op1 = tf.add(c2, c3)
op2 = tf.multiply(c2, c3)
print('op1 : ', op1)
print('op2 : ', op2)

op1 :  Tensor("Add:0", shape=(), dtype=float32)
op2 :  Tensor("Mul:0", shape=(), dtype=float32)

print('run(op1) : ', tfs.run(op1))
print('run(op2) : ', tfs.run(op2))

run(op1) :  13.0
run(op2) :  42.0

Placeholders

p1 = tf.placeholder(tf.float32)
p2 = tf.placeholder(tf.float32)
print('p1 : ', p1)
print('p2 : ', p2)

p1 :  Tensor("Placeholder:0", dtype=float32)
p2 :  Tensor("Placeholder_1:0", dtype=float32)

op4 = p1 * p2  # shorthand for tf.multiply(p1, p2)

print('run(op4,{p1:2.0, p2:3.0}) : ',tfs.run(op4,{p1:2.0, p2:3.0}))

run(op4,{p1:2.0, p2:3.0}) :  6.0

print('run(op4,feed_dict = {p1:3.0, p2:4.0}) : ',
      tfs.run(op4, feed_dict={p1: 3.0, p2: 4.0}))

run(op4,feed_dict = {p1:3.0, p2:4.0}) :  12.0

print('run(op4,feed_dict={p1:[2.0,3.0,4.0], p2:[3.0,4.0,5.0]}):',
      tfs.run(op4, feed_dict={p1: [2.0, 3.0, 4.0], p2: [3.0, 4.0, 5.0]}))

run(op4,feed_dict = {p1:[2.0,3.0,4.0], p2:[3.0,4.0,5.0]}) :  [  6.  12.  20.]

Creating Tensors from Existing Objects

0-Dimensional Tensors (Scalars)

tf_t = tf.convert_to_tensor(5.0, dtype=tf.float64)

print('tf_t : ', tf_t)
print('run(tf_t) : \n', tfs.run(tf_t))

tf_t :  Tensor("Const_1:0", shape=(), dtype=float64)
run(tf_t) : 
 5.0

1-Dimensional Tensors (Vectors)

a1dim = np.array([1, 2, 3, 4, 5.99])
print("a1dim Shape : ", a1dim.shape)

tf_t = tf.convert_to_tensor(a1dim, dtype=tf.float64)

print('tf_t : ', tf_t)
print('tf_t[0] : ', tf_t[0])
print('tf_t[0] : ', tf_t[2])
print('run(tf_t) : \n', tfs.run(tf_t))

a1dim Shape :  (5,)
tf_t :  Tensor("Const_2:0", shape=(5,), dtype=float64)
tf_t[0] :  Tensor("strided_slice:0", shape=(), dtype=float64)
tf_t[0] :  Tensor("strided_slice_1:0", shape=(), dtype=float64)
run(tf_t) : 
 [ 1.    2.    3.    4.    5.99]

2-Dimensional Tensors (Matrices)

a2dim = np.array([(1, 2, 3, 4, 5.99),
                  (2, 3, 4, 5, 6.99),
                  (3, 4, 5, 6, 7.99)
                  ])
print("a2dim Shape : ", a2dim.shape)

tf_t = tf.convert_to_tensor(a2dim, dtype=tf.float64)

print('tf_t : ', tf_t)
print('tf_t[0][0] : ', tf_t[0][0])
print('tf_t[1][2] : ', tf_t[1][2])
print('run(tf_t) : \n', tfs.run(tf_t))

a2dim Shape :  (3, 5)
tf_t :  Tensor("Const_3:0", shape=(3, 5), dtype=float64)
tf_t[0][0] :  Tensor("strided_slice_3:0", shape=(), dtype=float64)
tf_t[1][2] :  Tensor("strided_slice_5:0", shape=(), dtype=float64)
run(tf_t) : 
 [[ 1.    2.    3.    4.    5.99]
 [ 2.    3.    4.    5.    6.99]
 [ 3.    4.    5.    6.    7.99]]

3-Dimensional Tensors

a3dim = np.array([[[1, 2],
                   [3, 4]
                   ],
                  [[5, 6],
                   [7, 8]
                   ]
                  ])
print("a3dim Shape : ", a3dim.shape)

tf_t = tf.convert_to_tensor(a3dim, dtype=tf.float64)

print('tf_t : ', tf_t)
print('tf_t[0][0][0] : ', tf_t[0][0][0])
print('tf_t[1][1][1] : ', tf_t[1][1][1])
print('run(tf_t) : \n', tfs.run(tf_t))

a3dim Shape :  (2, 2, 2)
tf_t :  Tensor("Const_4:0", shape=(2, 2, 2), dtype=float64)
tf_t[0][0][0] :  Tensor("strided_slice_8:0", shape=(), dtype=float64)
tf_t[1][1][1] :  Tensor("strided_slice_11:0", shape=(), dtype=float64)
run(tf_t) : 
 [[[ 1.  2.]
  [ 3.  4.]]

 [[ 5.  6.]
  [ 7.  8.]]]

Variables

# Assume Linear Model y = w * x + b
# Define model parameters
w = tf.Variable([.3], tf.float32)
b = tf.Variable([-.3], tf.float32)
# Define model input and output
x = tf.placeholder(tf.float32)
y = w * x + b

print("w:", w)
print("x:", x)
print("b:", b)
print("y:", y)

w: <tf.Variable 'Variable:0' shape=(1,) dtype=float32_ref>
x: Tensor("Placeholder_2:0", dtype=float32)
b: <tf.Variable 'Variable_1:0' shape=(1,) dtype=float32_ref>
y: Tensor("add:0", dtype=float32)

# initialize and print the variable y
tf.global_variables_initializer().run()
print('run(y,{x:[1,2,3,4]}) : ', tfs.run(y, {x: [1, 2, 3, 4]}))

run(y,{x:[1,2,3,4]}) :  [ 0.          0.30000001  0.60000002  0.90000004]

Creating Tensors from Library Functions

a = tf.zeros((100,))
print(tfs.run(a))

[ 0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]

Close the interactive session

tfs.close()

Computation Graphs

Building and Running simple computation graph

# Assume Linear Model y = w * x + b
# Define model parameters
w = tf.Variable([.3], tf.float32)
b = tf.Variable([-.3], tf.float32)
# Define model input and output
x = tf.placeholder(tf.float32)
y = w * x + b
output = 0

with tf.Session() as tfs:
    # initialize and print the variable y
    tf.global_variables_initializer().run()
    output = tfs.run(y, {x: [1, 2, 3, 4]})
print('output : ', output)

output :  [ 0.          0.30000001  0.60000002  0.90000004]

Graph on Compute Devices

from tensorflow.python.client import device_lib

print(device_lib.list_local_devices())

[name: "/cpu:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 12445270569278384213
, name: "/gpu:0"
device_type: "GPU"
memory_limit: 25628672
locality {
  bus_id: 1
}
incarnation: 5284662930836416221
physical_device_desc: "device: 0, name: GeForce GT 750M, pci bus id: 0000:01:00.0"
]

tf.reset_default_graph()

# Define model parameters
w = tf.get_variable(name='w', initializer=[.3], dtype=tf.float32)
b = tf.get_variable(name='b', initializer=[-.3], dtype=tf.float32)
# Define model input and output
x = tf.placeholder(name='x', dtype=tf.float32)
y = w * x + b

config = tf.ConfigProto()
config.log_device_placement = True

with tf.Session(config=config) as tfs:
    # initialize and print the variable y
    tfs.run(tf.global_variables_initializer())
    print('output', tfs.run(y, {x: [1, 2, 3, 4]}))

output from GPU: [ 0.          0.30000001  0.60000002  0.90000004]

tf.reset_default_graph()

with tf.device('/device:CPU:0'):
    # Define model parameters
    w = tf.get_variable(name='w', initializer=[.3], dtype=tf.float32)
    b = tf.get_variable(name='b', initializer=[-.3], dtype=tf.float32)
    # Define model input and output
    x = tf.placeholder(name='x', dtype=tf.float32)
    y = w * x + b

config = tf.ConfigProto()
config.log_device_placement = True

with tf.Session(config=config) as tfs:
    # initialize and print the variable y
    tfs.run(tf.global_variables_initializer())
    print('output', tfs.run(y, {x: [1, 2, 3, 4]}))

tf.reset_default_graph()

with tf.device('/device:CPU:0'):
    # Define model parameters
    w = tf.get_variable(name='w', initializer=[.3], dtype=tf.float32)
    b = tf.get_variable(name='b', initializer=[-.3], dtype=tf.float32)
    # Define model input and output
    x = tf.placeholder(name='x', dtype=tf.float32)
with tf.device('/device:GPU:0'):
    y = w * x + b

config = tf.ConfigProto()
config.log_device_placement = True

with tf.Session(config=config) as tfs:
    # initialize and print the variable y
    tfs.run(tf.global_variables_initializer())
    print('output', tfs.run(y, {x: [1, 2, 3, 4]}))

Executing Graph g as Default

g = tf.Graph()
output = 0

# Assume Linear Model y = w * x + b

with g.as_default():
    # Define model parameters
    w = tf.Variable([.3], tf.float32)
    b = tf.Variable([-.3], tf.float32)
    # Define model input and output
    x = tf.placeholder(tf.float32)
    y = w * x + b

with tf.Session(graph=g) as tfs:
    # initialize and print the variable y
    tf.global_variables_initializer().run()
    output = tfs.run(y, {x: [1, 2, 3, 4]})

print('output : ', output)

output :  [ 0.          0.30000001  0.60000002  0.90000004]

TensorBoard

import tensorflow as tf
import numpy as np

# Assume Linear Model y = w * x + b
# Define model parameters
w = tf.Variable([.3], name='w', dtype=tf.float32)
b = tf.Variable([-.3], name='b', dtype=tf.float32)
# Define model input and output
x = tf.placeholder(name='x', dtype=tf.float32)
y = w * x + b

with tf.Session() as tfs:
    tf.global_variables_initializer().run()
    writer = tf.summary.FileWriter('tflogs', tfs.graph)
    print('run(y,{x:3}) : ', tfs.run(y, feed_dict={x: 3}))

run(y,{x:3}) :  [ 0.60000002]

# execute: tensorboard --logdir=logs

results matching ""

    No results matching ""