tf2.0

import tensorflow as tf
tf.__version__

!pip install tensorflow

#张量
#常量
a = tf.constant(2, name='a')
b = tf.constant(3, name='b')
x = tf.add(a, b)
print(x)
print(a+b)

a.get_shape()

a.numpy()

#变量
s = tf.Variable(2, name="scalar") 
m = tf.Variable([[0, 1], [2, 3]], name="matrix") 
W = tf.Variable(tf.zeros([784,10]))
s.assign(3)
s.assign_add(3)

class MyModule(tf.Module):
    def __init__(self):
        self.v0 = tf.Variable(1.0)
        self.vs = [tf.Variable(x) for x in range(10)]

m = MyModule()
m.variables

#tf.data
dataset = tf.data.Dataset.from_tensors([1,2,3,4,5])
for element in dataset:
  print(element.numpy())
  it = iter(dataset)
print(next(it).numpy())

dataset = tf.data.Dataset.from_tensor_slices([1,2,3,4,5])
for element in dataset:
  print(element.numpy())
it = iter(dataset)
print(next(it).numpy())

features = tf.data.Dataset.from_tensor_slices([1,2,3,4,5])
labels = tf.data.Dataset.from_tensor_slices([6,7,8,9,10])
dataset = tf.data.Dataset.zip((features,labels))
for element in dataset3:
  print(element)
  
inc_dataset = tf.data.Dataset.range(100)   
dec_dataset = tf.data.Dataset.range(0, -100, -1) 
dataset = tf.data.Dataset.zip((inc_dataset, dec_dataset))
batched_dataset = dataset.batch(4)  

for batch in batched_dataset.take(4):
  print([arr.numpy() for arr in batch])

shuffle_dataset = dataset.shuffle(buffer_size=10)
for element in shuffle_dataset:
  print(element)
  
shuffle_dataset = dataset.shuffle(buffer_size=100)
for element in shuffle_dataset:
  print(element)