1、tensorflow常用函数
TensorFlow 将图形定义转换成分布式执行的操作, 以充分利用可用的计算资源(如CPU 或GPU。一般你不需要显式指定使用CPU 还是GPU, TensorFlow 能自动检测。如果检测到GPU, TensorFlow 会尽可能地利用到的第一个GPU 来执行操作.
并行计算能让代价大的算法计算加速执行，TensorFlow也在实现上对复杂操作进行了有效的改进。大部分核相关的操作都是设备相关的实现，比如GPU。下面是一些重要的操作/核：
操作组操作
Maths Add, Sub, Mul, Div, Exp, Log, Greater, Less, Equal
Array Concat, Slice, Split, Constant, Rank, Shape, Shuffle
Matrix MatMul, MatrixInverse, MatrixDeterminant
Neuronal Network SoftMax, Sigmoid, ReLU, Convolution2D, MaxPool
Checkpointing Save, Restore
Queues and syncronizations Enqueue, Dequeue, MutexAcquire, MutexRelease
Flow control Merge, Switch, Enter, Leave, NextIteration
TensorFlow的算术操作如下：
操作描述
tf.add(x, y, name=None) 求和
tf.sub(x, y, name=None) 减法
tf.mul(x, y, name=None) 乘法
tf.div(x, y, name=None) 除法
tf.abs(x, name=None) 求绝对值
<(x, name=None) 取负(y = -x).
tf.sign(x, name=None) 返回符号y = sign(x) = -1 if x < 0; 0 if x == 0; 1 if x > 0.
tf.inv(x, name=None) 取反
tf.square(x, name=None) 计算平方(y = x * x = x^2).
# ‘a’ is [0.9, 2.5, 2.3, -4.4]
tf.sqrt(x, name=None) 开根号(y = \sqrt{x} = x^{1/2}).
tf.pow(x, y, name=None) 幂次方
# tensor ‘x’ is [[2, 2], [3, 3]]
# tensor ‘y’ is [[8, 16], [2, 3]]
tf.pow(x, y) ==> [[256, 65536], [9, 27]]input命令
tf.log(x, name=None) 计算log，一个输入计算e的ln，两输入以第二输入为底tf.maximum(x, y, name=None) 返回最大值(x > y ? x : y)
tf.minimum(x, y, name=None) 返回最小值(x < y ? x : y)
tf.sin(x, name=None) 三角函数sine
tf.tan(x, name=None) 三角函数tan
tf.atan(x, name=None) 三角函数ctan
张量操作Tensor Transformations
•数据类型转换Casting
操作描述
tf.string_to_number
(string_tensor, out_type=None, name=None) 字符串转为数字
<_double(x, name=’ToDouble’) 转为64位浮点类型–float64
<_float(x, name=’ToFloat’) 转为32位浮点类型–float32
<_int32(x, name=’ToInt32’) 转为32位整型–int32
<_int64(x, name=’ToInt64’) 转为64位整型–int64
tf.cast(x, dtype, name=None) 将x或者x.values转换为dtype
# tensor a is [1.8, 2.2], dtype=tf.float
tf.cast(a, tf.int32) ==> [1, 2] # dtype=tf.int32
•形状操作Shapes and Shaping
操作描述
tf.shape(input, name=None) 返回数据的shape
# ‘t’ is [[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]] shape(t) ==> [2, 2, 3]
tf.size(input, name=None) 返回数据的元素数量
# ‘t’ is [[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]]]
size(t) ==> 12
tf.rank(input, name=None) 返回tensor的rank
注意：此rank不同于矩阵的rank，
tensor的rank表示一个tensor需要的索引数目来唯一表示任何一个元素也就是通常所说的“order”, “degree”或”ndims”
#’t’ is [[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]]
# shape of tensor ‘t’ is [2, 2, 3]
rank(t) ==> 3
# tensor ‘t’ is [1, 2, 3, 4, 5, 6, 7, 8, 9]
# tensor ‘t’ has shape [9]
reshape(t, [3, 3]) ==>
[[1, 2, 3],
[4, 5, 6],
[7, 8, 9]]
#如果shape有元素[-1]，表示在该维度打平至一维# -1 将自动推导得为9:
reshape(t, [2, -1]) ==>
[[1, 1, 1, 2, 2, 2, 3, 3, 3],
[4, 4, 4, 5, 5, 5, 6, 6, 6]]
#该操作要求-1-input.dims()
# ‘t’ is a tensor of shape [2]
shape(expand_dims(t, 0)) ==> [1, 2]
shape(expand_dims(t, 1)) ==> [2, 1]
shape(expand_dims(t, -1)) ==> [2, 1] <= dim <= input.dims()
•切片与合并（Slicing and Joining）
操作描述
tf.slice(input_, begin, size, name=None) 对tensor进行切片操作
其中size[i] = input.dim_size(i) - begin[i]
该操作要求0 <= begin[i] <= begin[i] + size[i] <= Di for i in [0, n]
#’input’ is
#[[[1, 1, 1], [2, 2, 2]],[[3, 3, 3], [4, 4, 4]],[[5, 5, 5], [6, 6, 6]]] tf.slice(input, [1, 0, 0], [1, 1, 3]) ==> [[[3, 3, 3]]]
tf.slice(input, [1, 0, 0], [1, 2, 3]) ==>
[[[3, 3, 3],
[4, 4, 4]]]
tf.slice(input, [1, 0, 0], [2, 1, 3]) ==>
[[[3, 3, 3]],
[[5, 5, 5]]]
tf.split(split_dim, num_split, value, name=’split’) 沿着某一维度将tensor分离为num_split tensors # ‘value’ is a tensor with shape [5, 30]
# Split ‘value’ into 3 tensors along dimension 1 split0, split1, split2 = tf.split(1, 3, value)
tf.shape(split0) ==> [5, 10]
t1 = [[1, 2, 3], [4, 5, 6]]
t2 = [[7, 8, 9], [10, 11, 12]]
等同于tf.pack(tensors, axis=axis)
tf.pack(values, axis=0, name=’pack’) 将一系列rank-R的tensor打包为一个rank-(R+1)的tensor # ‘x’ is [1, 4], ‘y’ is [2, 5], ‘z’ is [3, 6]
pack([x, y, z]) => [[1, 4], [2, 5], [3, 6]]
# 沿着第一维pack
pack([x, y, z], axis=1) => [[1, 2, 3], [4, 5, 6]]
等价于tf.pack([x, y, z]) = np.asarray([x, y, z])
其中dim为列表，元素为bool型，size等于rank(tensor) # tensor ‘t’ is
[[[[ 0, 1, 2, 3],
#[ 4, 5, 6, 7],
#[ 8, 9, 10, 11]],
#[[12, 13, 14, 15],
#[16, 17, 18, 19],
#[20, 21, 22, 23]]]]
# tensor ‘t’ shape is [1, 2, 3, 4]
# ‘dims’ is [False, False, False, True]
reverse(t, dims) ==>
[[[[ 3, 2, 1, 0],
[ 7, 6, 5, 4],
[ 11, 10, 9, 8]],
[[15, 14, 13, 12],
[19, 18, 17, 16],
[23, 22, 21, 20]]]]
按照列表perm的维度排列调换tensor顺序，如为定义，则perm为(n-1…0)
# ‘x’ is [[1 2 3],[4 5 6]]
# Equivalently
name=None)
合并索引indices所指示params中的切片
<_hot
(indices, depth, on_value=None, off_value=None,  axis=None, dtype=None, name=None) indices = [0, 2, -1, 1] depth = 3
on_value = 5.0
off_value = 0.0
axis = -1
#Then output is [4 x 3]:  output =
[5.0 0.0 0.0] // one_hot(0)  [0.0 0.0 5.0] // one_hot(2)  [0.0 0.0 0.0] // one_hot(-1)  [0.0 5.0 0.0] // one_hot(1)
矩阵相关运算
操作描述
tf.diag(diagonal, name=None) 返回一个给定对角值的对角tensor # ‘diagonal’ is [1, 2, 3, 4]
tf.diag(diagonal) ==>
[[1, 0, 0, 0]
[0, 2, 0, 0]
[0, 0, 3, 0]
[0, 0, 0, 4]]
tf.diag_part(input, name=None) 功能与上面相反
按照列表perm的维度排列调换tensor顺序，如为定义，则perm为(n-1…0)
# ‘x’ is [[1 2 3],[4 5 6]]
# Equivalently
tf.matmul(a, b, transpose_a=False,
transpose_b=False, a_is_sparse=False,  矩阵相乘

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。