TensorFlow Learning Notes #1

基礎

1.Tensors are the primary data structure that TensorFlow uses to operate on the computational graph.

#Create fixed tensors
zero_tsr = tf.zeros([row_dim, col_dim])
ones_tsr = tf.ones([row_dim, col_dim])
filled_tsr = tf.fill([row_dim, col_dim], 42)
constant_tsr =
 
 tf.constant([1,2,3])

#Create similar shape of tensors
zeros_sim = tf.zeros_like(constant_tsr)
zeros_sim = tf.ones_like(constant_tsr)

#Create sequence tensors
linear_tsr = tf.linspace(start=0.0, stop=1.0, num=3.0)
integer_seq_tsr = tf.range(start=6, limit=15, delta=3)

#Create random tensors
randunif_tsr =
 
 tf.random_uniform([row.dim, col_dim], minval=0, maxval=1)
randnorm_tsr = tf.random_normal([row_dim, col_dim], mean=0.0, stddev=1.0)
runcnorm_tsr = tf.truncated_normal([row_dim, col_dim], mean=0.0, stddev=1.0)
shuffled_output = tf.random_shuffle(input_tensor)
cropped_output = tf.random_crop(input_tensor, crop_size)
 

cropped_image = tf.random_crop(my_image, [height/2, width/2, 3])

2.Variables are the parameters of the algorithm and TensorFlow keeps track of how to change these to optimize the algorithm. Placeholders are objects that allow you to feed in data of a specific type and shape and depend on the results of the computational graph, such as the expected outcome of a computation.

#Create and initialize a variable
my_var = tf.Variable(tf.zeros([2, 3]))
sess = tf.Session()
initialize_op = tf.global_variables_initializer()
sess.run(initialize_op)

#Declar a placeholder and perform operations on it
sess = tf.Session()
x = tf.placeholder(tf.float32, shape=[2,2])
y = tf.identity(x)
x_vals = np.random.rand(2,2)
sess.run(y, feed_dict={x: x_vals})

3.Matrices

#Create matrices
import tensorflow as tf
sess = tf.Session()
identity_matrix = tf.diag([1.0, 1.0, 1.0])
A = tf.truncated_normal([2, 3])
B = tf.fill([2,3], 5.0)
C = tf.random_uniform([3,2])
D = tf.convert_to_tensor(np.array([[1., 2., 3.], [-3., -7., -1.], [0., 5., -2.]]))
print(sess.run(identity_matrix))

#Addition, subtraction and multiplication
print(sess.run(A+B))
print(sess.run(B-B))
print(sess.run(tf.matmul(B, identity_matrix)))

#Transpose, inverse and determinant
print(sess.run(tf.transpose(C)))
print(sess.run(tf.matrix_inverse(D)))
print(sess.run(tf.matrix_determinant(D)))

#Decompositoins
print(sess.run(tf.cholesky(identity_matrix)))

#Eigenvalues and eigenvectors
print(sess.run(tf.self_adjoint_eig(D))

4.Operations

5.Activation Functions

#Create rectifier linear unit, ReLU
import tensorflow as tf
sess = tf.Session()
print(sess.run(tf.nn.relu([-3., 3., 10.])))

#Create ReLU6 = min(max(0,x),6)
print(sess.run(tf.nn.relu6([-3., 3., 10.])))

#Create sigmoid function = 1/(1+exp(-x))
print(sess.run(tf.nn.sigmoid([-1., 0., 1.])))

#Create tanh function = ((exp(x)-exp(-x))/(exp(x)+exp(-x))
print(sess.run(tf.nn.tanh([-1., 0., 1.])))

#Create softsign function = x/(abs(x)+1）
print(sess.run(tf.nn.softsign([-1., 0., -1.])))

#Create softplus function = log(exp(x)+1)
print(sess.run(tf.nn.softplus([-1., 0., -1.])))

#Create Exponential Linear Unit, ELU function = (exp(x)+1) if x<0 else x
print(sess.run(tf.nn.elu([-1., 0., -1.])))

6.Data Sources
How to access datasets through TensorFlow and Python

#Iris data
from sklearn import datasets
iris = datasets.load_iris()
print(len(iris.data))
print(len(iris.target))
print(iris.target[0])
print(set(iris.target))

#Birth weight data
import requests
birthdata_url = 'http://faculty.washington.edu/heagerty/Courses/b513/WEB2002/datasets/lowbwt.dat'
birth_file = requests.get(birthdata_url)
birth_header = ['ID','LOW','AGE','LWT','RACE','SMOKE','PTL','HT','UI','FTV','BWT']
birth_data = [[float(x) for x in y.split(' ') if len(x) >=1] for y in birth_file.text.split('\n') if len(y) >=1]
print(len(birth_data))
print(len(birth_data[0]))

#Boston Housing data
import requests
housing_url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/housing/housing.data'
housing_header = ['CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX', 'PTRATIO', 'B', 'LSTAT', 'MEDV']
housing_file = requests.get(housing_url)
housing_data = [[float(x) for x in y.split(' ') if len(x)>=1] for y in housing_file.text.split('\n') if len(y)>=1]
print(len(housing_data))
print(len(housing_data[0]))

#MNIST handwriting data: (Mixed National Instituite of Standards and Technology)
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
print(len(mnist.train.images))
print(len(mnist.test.images))
print(len(mnist.validation.images))
print(mnist.train.labels[1,:])

#Spam-ham text data
import requests
import io
from zipfile import ZipFile
zip_url = 'http://archive.ics.uci.edu/ml/machine-learning-databases/00228/smsspamcollection.zip'
r = requests.get(zip_url)
z = ZipFile(io.BytesIO(r.content))
file = z.read('SMSSpamCollection')
text_data = file.decode()
text_data = text_data.encode('ascii', errors='ignore')
text_data = text_data.decode().split('\n')
text_data = [x.split('\t') for x in text_data if len(x) >=1]
[text_data_target, text_data_train] = [list(x) for x in zip(*text_data)]
print(len(text_data_train))
print(set(text_data_target))
print(text_data_train[1])

#Movie review data
import requests
import io
import tarfile
movie_data_url = 'http://www.cs.cornell.edu/people/pabo/movie-review-data/rt-polaritydata.tar.gz'
r = requests.get(movie_data_url)
stream_data = io.BytesIO(r.content)
tmp = io.BytesIO()
while True:
    s = stream_data.read(16384)
    if not s:
        break
    tmp.write(s)
stream_data.close()
tmp.seek(0)
#Extract tar file
tar_file = tarfile.open(fileobj=tmp, mode='r:gz')
pos = tar_file.extractfile('rt-polaritydata/rt-polarity.pos')
neg = tar_file.extractfile('rt-polaritydata/rt-polarity.neg')
#Save pos/neg reviews (Also deal with encoding)
pos_data = []
for line in pos:
    pos_data.append(line.decode("ISO'-8859-1").encode('ascii',errors='ignore').decode())
neg_data = []
for line in neg:
    neg_data.append(line.decode("ISO'-8859-1").encode('ascii',errors='ignore').decode())
tar_file.close()
print(len(pos_data))
print(len(neg_data))
print(neg_data[0])

#The works of Shakespeare text data
import requests
shakespeare_url = 'http://www.gutenberg.org/cache/epub/100/pg100.txt'
response = requests.get(shakespeare_url)
shakespeare_file = response.content
shakespeare_text = shakespeare_file.decode('utf-8')
shakespeare_text = shakespeare_text[7675:]
print(len(shakespeare_text))

#English-German sentence translation data
import requests
import io
from zipfile import ZipFile
sentence_url = 'http://www.manythings.org/anki/deu-eng.zip'
r = requests.get(sentence_url)
z = ZipFile(io.BytesIO(r.content))
file = z.read('deu.txt')
eng_ger_data = file.decode()
eng_ger_data = eng_ger_data.encode('ascii', errors='ignore')
eng_ger_data = eng_ger_data.decode().split('\n')
eng_ger_data = [x.split('\t') for x in eng_ger_data if len(x) >=1]
[english_sentence, german_sentence] = [list(x) for x in zip(*eng_ger_data)]
print(len(english_sentence))
print(len(german_sentence))
print(eng_ger_data[10])

7.參考資料

• 《TensorFlow Machine Learning Cookbook》：https://github.com/PacktPublishing/TensorFlow-Machine-Learning-Cookbook
• TensorFlow API：https://www.tensorflow.org/api_docs/python/tf
• Google提供的TensorFlow測試環境：https://colab.research.google.com