正文
TensorFlow-相关 API（学习笔记 ）
小程序：扫一扫查出行【扫一扫了解最新限行尾号】
复制小程序
1.tf.nn.conv2d
conv2d(
input,
filter,
strides,
padding,
use_cudnn_on_gpu=True,
data_format='NHWC',
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| input | 是 | tensor | 是一个 4 维的 tensor，即 [ batch, in_height, in_width, in_channels ]（若 input 是图像，[ 训练时一个 batch 的图片数量, 图片高度, 图片宽度, 图像通道数 ]） |
| filter | 是 | tensor | 是一个 4 维的 tensor，即 [ filter_height, filter_width, in_channels, out_channels ]（若 input 是图像，[ 卷积核的高度，卷积核的宽度，图像通道数，卷积核个数 ]）,filter 的 in_channels 必须和 input 的 in_channels 相等 |
| strides | 是 | 列表 | 长度为 4 的 list，卷积时候在 input 上每一维的步长，一般 strides[0] = strides[3] = 1 |
| padding | 是 | string | 只能为 " VALID "，" SAME " 中之一，这个值决定了不同的卷积方式。VALID 丢弃方式；SAME：补全方式 |
| use_cudnn_on_gpu | 否 | bool | 是否使用 cudnn 加速，默认为 true |
| data_format | 否 | string | 只能是 " NHWC ", " NCHW "，默认 " NHWC " |
| name | 否 | string | 运算名称 |
创建conv2d.py
import tensorflow as tfa = tf.constant([1,1,1,0,0,0,1,1,1,0,0,0,1,1,1,0,0,1,1,0,0,1,1,0,0],dtype=tf.float32,shape=[1,5,5,1])
b = tf.constant([1,0,1,0,1,0,1,0,1],dtype=tf.float32,shape=[3,3,1,1])
c = tf.nn.conv2d(a,b,strides=[1, 2, 2, 1],padding='VALID')
d = tf.nn.conv2d(a,b,strides=[1, 2, 2, 1],padding='SAME')
with tf.Session() as sess:
print ("c shape:")
print (c.shape)
print ("c value:")
print (sess.run(c))
print ("d shape:")
print (d.shape)
print ("d value:")
print (sess.run(d))
执行结果：
c shape:
(1, 2, 2, 1)
c value:
[[[[ 4.]
[ 4.]] [[ 2.]
[ 4.]]]]
d shape:
(1, 3, 3, 1)
d value:
[[[[ 2.]
[ 3.]
[ 1.]] [[ 1.]
[ 4.]
[ 3.]] [[ 0.]
[ 2.]
[ 1.]]]]
2.tf.nn.relu
relu(
features,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| features | 是 | tensor | 是以下类型float32, float64, int32, int64, uint8, int16, int8, uint16, half |
| name | 否 | string | 运算名称 |
创建源文件 relu.py
import tensorflow as tfa = tf.constant([1,-2,0,4,-5,6])
b = tf.nn.relu(a)
with tf.Session() as sess:
print (sess.run(b))
执行结果：
[1 0 0 4 0 6]
3.tf.nn.max_pool
max_pool(
value,
ksize,
strides,
padding,
data_format='NHWC',
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| value | 是 | tensor | 4 维的张量，即 [ batch, height, width, channels ]，数据类型为 tf.float32 |
| ksize | 是 | 列表 | 池化窗口的大小，长度为 4 的 list，一般是 [1, height, width, 1]，因为不在 batch 和 channels 上做池化，所以第一个和最后一个维度为 1 |
| strides | 是 | 列表 | 池化窗口在每一个维度上的步长，一般 strides[0] = strides[3] = 1 |
| padding | 是 | string | 只能为 " VALID "，" SAME " 中之一，这个值决定了不同的池化方式。VALID 丢弃方式；SAME：补全方式 |
| data_format | 否 | string | 只能是 " NHWC ", " NCHW "，默认" NHWC " |
| name | 否 | string | 运算名称 |
创建源文件max_pool.py
import tensorflow as tfa = tf.constant([1,3,2,1,2,9,1,1,1,3,2,3,5,6,1,2],dtype=tf.float32,shape=[1,4,4,1])
b = tf.nn.max_pool(a,ksize=[1, 2, 2, 1],strides=[1, 2, 2, 1],padding='VALID')
c = tf.nn.max_pool(a,ksize=[1, 2, 2, 1],strides=[1, 2, 2, 1],padding='SAME')
with tf.Session() as sess:
print ("b shape:")
print (b.shape)
print ("b value:")
print (sess.run(b))
print ("c shape:")
print (c.shape)
print ("c value:")
print (sess.run(c))
执行结果：
b shape:
(1, 2, 2, 1)
b value:
[[[[ 9.]
[ 2.]] [[ 6.]
[ 3.]]]]
c shape:
(1, 2, 2, 1)
c value:
[[[[ 9.]
[ 2.]] [[ 6.]
[ 3.]]]]
4.tf.nn.dropout
dropout(
x,
keep_prob,
noise_shape=None,
seed=None,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| x | 是 | tensor | 输出元素是 x 中的元素以 keep_prob 概率除以 keep_prob，否则为 0 |
| keep_prob | 是 | scalar Tensor | dropout 的概率，一般是占位符 |
| noise_shape | 否 | tensor | 默认情况下，每个元素是否 dropout 是相互独立。如果指定 noise_shape，若 noise_shape[i] == shape(x)[i]，该维度的元素是否 dropout 是相互独立，若 noise_shape[i] != shape(x)[i] 该维度元素是否 dropout 不相互独立，要么一起 dropout 要么一起保留 |
| seed | 否 | 数值 | 如果指定该值，每次 dropout 结果相同 |
| name | 否 | string | 运算名称 |
创建源文件dropout.py
import tensorflow as tfa = tf.constant([1,2,3,4,5,6],shape=[2,3],dtype=tf.float32)
b = tf.placeholder(tf.float32)
c = tf.nn.dropout(a,b,[2,1],1)
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
print (sess.run(c,feed_dict={b:0.75}))
执行结果：
[[ 0. 0. 0. ]
[ 5.33333349 6.66666651 8. ]]
5. tf.nn.sigmoid_cross_entropy_with_logits
sigmoid_cross_entropy_with_logits(
_sentinel=None,
labels=None,
logits=None,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| _sentinel | 否 | None | 没有使用的参数 |
| labels | 否 | Tensor | type, shape 与 logits相同 |
| logits | 否 | Tensor | type 是 float32 或者 float64 |
| name | 否 | string | 运算名称 |
创建源文件sigmoid_cross_entropy_with_logits.py
import tensorflow as tf
x = tf.constant([1,2,3,4,5,6,7],dtype=tf.float64)
y = tf.constant([1,1,1,0,0,1,0],dtype=tf.float64)
loss = tf.nn.sigmoid_cross_entropy_with_logits(labels = y,logits = x)
with tf.Session() as sess:
print (sess.run(loss))
执行结果：
[ 3.13261688e-01 1.26928011e-01 4.85873516e-02 4.01814993e+00
5.00671535e+00 2.47568514e-03 7.00091147e+00]
6.tf.truncated_normal
truncated_normal(
shape,
mean=0.0,
stddev=1.0,
dtype=tf.float32,
seed=None,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| shape | 是 | 1 维整形张量或 array | 输出张量的维度 |
| mean | 否 | 0 维张量或数值 | 均值 |
| stddev | 否 | 0 维张量或数值 | 标准差 |
| dtype | 否 | dtype | 输出类型 |
| seed | 否 | 数值 | 随机种子，若 seed 赋值，每次产生相同随机数 |
| name | 否 | string | 运算名称 |
创建源文件 truncated_normal.py
import tensorflow as tf
initial = tf.truncated_normal(shape=[3,3], mean=0, stddev=1)
print(tf.Session().run(initial))
执行结果：
[[ 0.18815269 -0.4689253 0.63908994]
[ 0.01734953 -0.46975166 -0.25023392]
[ 1.12803638 -1.84143591 0.15422213]]
7.tf.constant
constant(
value,
dtype=None,
shape=None,
name='Const',
verify_shape=False
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| value | 是 | 常量数值或者 list | 输出张量的值 |
| dtype | 否 | dtype | 输出张量元素类型 |
| shape | 否 | 1 维整形张量或 array | 输出张量的维度 |
| name | 否 | string | 张量名称 |
| verify_shape | 否 | Boolean | 检测 shape 是否和 value 的 shape 一致，若为 Fasle，不一致时，会用最后一个元素将 shape 补全 |
创建源文件 constant.py
#!/usr/bin/pythonimport tensorflow as tf
import numpy as np
a = tf.constant([1,2,3,4,5,6],shape=[2,3])
b = tf.constant(-1,shape=[3,2])
c = tf.matmul(a,b)e = tf.constant(np.arange(1,13,dtype=np.int32),shape=[2,2,3])
f = tf.constant(np.arange(13,25,dtype=np.int32),shape=[2,3,2])
g = tf.matmul(e,f)
with tf.Session() as sess:
print (sess.run(a))
print ("##################################")
print (sess.run(b))
print ("##################################")
print (sess.run(c))
print ("##################################")
print (sess.run(e))
print ("##################################")
print (sess.run(f))
print ("##################################")
print (sess.run(g))
执行结果：
[[1 2 3]
[4 5 6]]
##################################
[[-1 -1]
[-1 -1]
[-1 -1]]
##################################
[[ -6 -6]
[-15 -15]]
##################################
[[[ 1 2 3]
[ 4 5 6]] [[ 7 8 9]
[10 11 12]]]
##################################
[[[13 14]
[15 16]
[17 18]] [[19 20]
[21 22]
[23 24]]]
##################################
[[[ 94 100]
[229 244]] [[508 532]
[697 730]]]
8.tf.placeholder
placeholder(
dtype,
shape=None,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| dtype | 是 | dtype | 占位符数据类型 |
| shape | 否 | 1 维整形张量或 array | 占位符维度 |
| name | 否 | string | 占位符名称 |
创建源文件 placeholder.py
#!/usr/bin/pythonimport tensorflow as tf
import numpy as npx = tf.placeholder(tf.float32,[None,3])
y = tf.matmul(x,x)
with tf.Session() as sess:
rand_array = np.random.rand(3,3)
print(sess.run(y,feed_dict={x:rand_array}))
执行结果：
[[ 0.64431196 0.68349576 0.57412398]
[ 0.84553117 1.64796805 0.7788316 ]
[ 0.84342241 0.8947317 0.8024016 ]]
9.tf.nn.bias_add 将偏差项 bias 加到 value 上面，可以看做是 tf.add 的一个特例，其中 bias 必须是一维的，并且维度和 value 的最后一维相同，数据类型必须和 value 相同。
bias_add(
value,
bias,
data_format=None,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| value | 是 | 张量 | 数据类型为 float, double, int64, int32, uint8, int16, int8, complex64, or complex128 |
| bias | 是 | 1 维张量 | 维度必须和 value 最后一维维度相等 |
| data_format | 否 | string | 数据格式，支持 ' NHWC ' 和 ' NCHW ' |
| name | 否 | string | 运算名称 |
创建源文件 bias_add.py
#!/usr/bin/pythonimport tensorflow as tf
import numpy as npa = tf.constant([[1.0, 2.0],[1.0, 2.0],[1.0, 2.0]])
b = tf.constant([2.0,1.0])
c = tf.constant([1.0])
sess = tf.Session()
print (sess.run(tf.nn.bias_add(a, b)))
#print (sess.run(tf.nn.bias_add(a,c))) error
print ("##################################")
print (sess.run(tf.add(a, b)))
print ("##################################")
print (sess.run(tf.add(a, c)))
执行结果：
[[ 3. 3.]
[ 3. 3.]
[ 3. 3.]]
##################################
[[ 3. 3.]
[ 3. 3.]
[ 3. 3.]]
##################################
[[ 2. 3.]
[ 2. 3.]
[ 2. 3.]]
10.tf.reduce_mean
reduce_mean(
input_tensor,
axis=None,
keep_dims=False,
name=None,
reduction_indices=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| input_tensor | 是 | 张量 | 输入待求平均值的张量 |
| axis | 否 | None、0、1 | None：全局求平均值；0：求每一列平均值；1：求每一行平均值 |
| keep_dims | 否 | Boolean | 保留原来的维度(例如不会从二维矩阵降为一维向量) |
| name | 否 | string | 运算名称 |
| reduction_indices | 否 | None | 和 axis 等价，被弃用 |
创建源文件 reduce_mean.py
#!/usr/bin/pythonimport tensorflow as tf
import numpy as npinitial = [[1.,1.],[2.,2.]]
x = tf.Variable(initial,dtype=tf.float32)
init_op = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init_op)
print(sess.run(tf.reduce_mean(x)))
print(sess.run(tf.reduce_mean(x,0))) #Column
print(sess.run(tf.reduce_mean(x,1))) #row
执行结果：
1.5
[ 1.5 1.5]
[ 1. 2.]
11.tf.squared_difference 计算张量 x、y 对应元素差平方
squared_difference(
x,
y,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| x | 是 | 张量 | 是 half, float32, float64, int32, int64, complex64, complex128 其中一种类型 |
| y | 是 | 张量 | 是 half, float32, float64, int32, int64, complex64, complex128 其中一种类型 |
| name | 否 | string | 运算名称 |
创建源文件 squared_difference.py
#!/usr/bin/pythonimport tensorflow as tf
import numpy as npinitial_x = [[1.,1.],[2.,2.]]
x = tf.Variable(initial_x,dtype=tf.float32)
initial_y = [[3.,3.],[4.,4.]]
y = tf.Variable(initial_y,dtype=tf.float32)
diff = tf.squared_difference(x,y)
init_op = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init_op)
print(sess.run(diff))
执行结果：
[[ 4. 4.]
[ 4. 4.]]
12.tf.square 计算张量对应元素平方
square(
x,
name=None
)
| 参数名 | 必选 | 类型 | 说明 |
|---|---|---|---|
| x | 是 | 张量 | 是 half, float32, float64, int32, int64, complex64, complex128 其中一种类型 |
| name | 否 | string | 运算名称 |
创建源文件 square.py
#!/usr/bin/python
import tensorflow as tf
import numpy as npinitial_x = [[1.,1.],[2.,2.]]
x = tf.Variable(initial_x,dtype=tf.float32)
x2 = tf.square(x)
init_op = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init_op)
print(sess.run(x2))
执行结果：
[[ 1. 1.]
[ 4. 4.]]
13.tf.Variable 维护图在执行过程中的状态信息，例如神经网络权重值的变化。
__init__(
initial_value=None,
trainable=True,
collections=None,
validate_shape=True,
caching_device=None,
name=None,
variable_def=None,
dtype=None,
expected_shape=None,
import_scope=None
)
| 参数名 | 类型 | 说明 |
|---|---|---|
| initial_value | 张量 | Variable 类的初始值，这个变量必须指定 shape 信息，否则后面 validate_shape 需设为 False |
| trainable | Boolean | 是否把变量添加到 collection GraphKeys.TRAINABLE_VARIABLES 中（collection 是一种全局存储，不受变量名生存空间影响，一处保存，到处可取） |
| collections | Graph collections | 全局存储，默认是 GraphKeys.GLOBAL_VARIABLES |
| validate_shape | Boolean | 是否允许被未知维度的 initial_value 初始化 |
| caching_device | string | 指明哪个 device 用来缓存变量 |
| name | string | 变量名 |
| dtype | dtype | 如果被设置，初始化的值就会按照这个类型初始化 |
| expected_shape | TensorShape | 要是设置了，那么初始的值会是这种维度 |
创建源文件Variable.py
#!/usr/bin/pythonimport tensorflow as tf
initial = tf.truncated_normal(shape=[10,10],mean=0,stddev=1)
W=tf.Variable(initial)
list = [[1.,1.],[2.,2.]]
X = tf.Variable(list,dtype=tf.float32)
init_op = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init_op)
print ("##################(1)################")
print (sess.run(W))
print ("##################(2)################")
print (sess.run(W[:2,:2]))
op = W[:2,:2].assign(22.*tf.ones((2,2)))
print ("###################(3)###############")
print (sess.run(op))
print ("###################(4)###############")
print (W.eval(sess)) #computes and returns the value of this variable
print ("####################(5)##############")
print (W.eval()) #Usage with the default session
print ("#####################(6)#############")
print (W.dtype)
print (sess.run(W.initial_value))
print (sess.run(W.op))
print (W.shape)
print ("###################(7)###############")
print (sess.run(X))
执行结果：
##################(1)################
[[-0.14252207 0.43376675 0.75065768 0.89276749 1.16391671 0.39532429
-0.56278807 -0.49753642 0.23130737 -0.51338279]
[-0.43028545 -1.24873769 -0.73239309 0.434468 -0.97399759 0.13766721
-0.6361087 -0.82712436 1.71831048 -0.44968474]
[-0.96064204 -0.83682173 0.26545268 0.22578485 0.65014762 -0.30830157
-1.57317054 -0.35661098 1.40849245 -0.37030414]
[-0.37272176 0.73461288 0.39292559 -1.40008056 -0.37535539 0.24140523
1.6811192 -0.48886588 1.15467834 0.61565816]
[-0.39579329 -0.23154807 -1.01895738 -0.95105737 1.24795806 -0.03846256
-1.71738017 -0.80132687 0.53553152 -0.06413679]
[-0.97320521 -0.24279226 1.36213648 1.56002438 -1.11646473 -0.35991025
0.91412318 0.97508883 -1.16207206 -0.68734062]
[ 0.49044254 -1.87386227 -0.70803815 -0.6591838 0.08034691 -1.24559033
-0.29389012 -0.2189652 -1.08279467 -0.0175346 ]
[-0.5608176 1.08259249 1.66278481 -0.33977437 0.42875817 0.55927169
0.76387608 0.37792665 0.85006535 1.05124724]
[ 1.75331545 -0.6333124 -0.10046791 -0.1780251 -1.31002116 1.90098214
0.84569824 -1.42502522 -0.67300171 0.68910873]
[-1.7385 -0.9806214 -0.32636395 -0.50020444 -0.53104508 -0.33903483
-0.35751811 -0.03737256 -1.26822579 -1.38264406]]
##################(2)################
[[-0.14252207 0.43376675]
[-0.43028545 -1.24873769]]
###################(3)###############
[[ 2.20000000e+01 2.20000000e+01 7.50657678e-01 8.92767489e-01
1.16391671e+00 3.95324290e-01 -5.62788069e-01 -4.97536421e-01
2.31307372e-01 -5.13382792e-01]
[ 2.20000000e+01 2.20000000e+01 -7.32393086e-01 4.34468001e-01
-9.73997593e-01 1.37667209e-01 -6.36108696e-01 -8.27124357e-01
1.71831048e+00 -4.49684739e-01]
[ -9.60642040e-01 -8.36821735e-01 2.65452683e-01 2.25784853e-01
6.50147617e-01 -3.08301568e-01 -1.57317054e+00 -3.56610984e-01
1.40849245e+00 -3.70304137e-01]
[ -3.72721761e-01 7.34612882e-01 3.92925590e-01 -1.40008056e+00
-3.75355393e-01 2.41405234e-01 1.68111920e+00 -4.88865882e-01
1.15467834e+00 6.15658164e-01]
[ -3.95793289e-01 -2.31548071e-01 -1.01895738e+00 -9.51057374e-01
1.24795806e+00 -3.84625569e-02 -1.71738017e+00 -8.01326871e-01
5.35531521e-01 -6.41367882e-02]
[ -9.73205209e-01 -2.42792264e-01 1.36213648e+00 1.56002438e+00
-1.11646473e+00 -3.59910250e-01 9.14123178e-01 9.75088835e-01
-1.16207206e+00 -6.87340617e-01]
[ 4.90442544e-01 -1.87386227e+00 -7.08038151e-01 -6.59183800e-01
8.03469121e-02 -1.24559033e+00 -2.93890119e-01 -2.18965203e-01
-1.08279467e+00 -1.75346043e-02]
[ -5.60817599e-01 1.08259249e+00 1.66278481e+00 -3.39774370e-01
4.28758174e-01 5.59271693e-01 7.63876081e-01 3.77926648e-01
8.50065351e-01 1.05124724e+00]
[ 1.75331545e+00 -6.33312404e-01 -1.00467913e-01 -1.78025097e-01
-1.31002116e+00 1.90098214e+00 8.45698237e-01 -1.42502522e+00
-6.73001707e-01 6.89108729e-01]
[ -1.73850000e+00 -9.80621397e-01 -3.26363951e-01 -5.00204444e-01
-5.31045079e-01 -3.39034826e-01 -3.57518107e-01 -3.73725556e-02
-1.26822579e+00 -1.38264406e+00]]
###################(4)###############
[[ 2.20000000e+01 2.20000000e+01 7.50657678e-01 8.92767489e-01
1.16391671e+00 3.95324290e-01 -5.62788069e-01 -4.97536421e-01
2.31307372e-01 -5.13382792e-01]
[ 2.20000000e+01 2.20000000e+01 -7.32393086e-01 4.34468001e-01
-9.73997593e-01 1.37667209e-01 -6.36108696e-01 -8.27124357e-01
1.71831048e+00 -4.49684739e-01]
[ -9.60642040e-01 -8.36821735e-01 2.65452683e-01 2.25784853e-01
6.50147617e-01 -3.08301568e-01 -1.57317054e+00 -3.56610984e-01
1.40849245e+00 -3.70304137e-01]
[ -3.72721761e-01 7.34612882e-01 3.92925590e-01 -1.40008056e+00
-3.75355393e-01 2.41405234e-01 1.68111920e+00 -4.88865882e-01
1.15467834e+00 6.15658164e-01]
[ -3.95793289e-01 -2.31548071e-01 -1.01895738e+00 -9.51057374e-01
1.24795806e+00 -3.84625569e-02 -1.71738017e+00 -8.01326871e-01
5.35531521e-01 -6.41367882e-02]
[ -9.73205209e-01 -2.42792264e-01 1.36213648e+00 1.56002438e+00
-1.11646473e+00 -3.59910250e-01 9.14123178e-01 9.75088835e-01
-1.16207206e+00 -6.87340617e-01]
[ 4.90442544e-01 -1.87386227e+00 -7.08038151e-01 -6.59183800e-01
8.03469121e-02 -1.24559033e+00 -2.93890119e-01 -2.18965203e-01
-1.08279467e+00 -1.75346043e-02]
[ -5.60817599e-01 1.08259249e+00 1.66278481e+00 -3.39774370e-01
4.28758174e-01 5.59271693e-01 7.63876081e-01 3.77926648e-01
8.50065351e-01 1.05124724e+00]
[ 1.75331545e+00 -6.33312404e-01 -1.00467913e-01 -1.78025097e-01
-1.31002116e+00 1.90098214e+00 8.45698237e-01 -1.42502522e+00
-6.73001707e-01 6.89108729e-01]
[ -1.73850000e+00 -9.80621397e-01 -3.26363951e-01 -5.00204444e-01
-5.31045079e-01 -3.39034826e-01 -3.57518107e-01 -3.73725556e-02
-1.26822579e+00 -1.38264406e+00]]
####################(5)##############
[[ 2.20000000e+01 2.20000000e+01 7.50657678e-01 8.92767489e-01
1.16391671e+00 3.95324290e-01 -5.62788069e-01 -4.97536421e-01
2.31307372e-01 -5.13382792e-01]
[ 2.20000000e+01 2.20000000e+01 -7.32393086e-01 4.34468001e-01
-9.73997593e-01 1.37667209e-01 -6.36108696e-01 -8.27124357e-01
1.71831048e+00 -4.49684739e-01]
[ -9.60642040e-01 -8.36821735e-01 2.65452683e-01 2.25784853e-01
6.50147617e-01 -3.08301568e-01 -1.57317054e+00 -3.56610984e-01
1.40849245e+00 -3.70304137e-01]
[ -3.72721761e-01 7.34612882e-01 3.92925590e-01 -1.40008056e+00
-3.75355393e-01 2.41405234e-01 1.68111920e+00 -4.88865882e-01
1.15467834e+00 6.15658164e-01]
[ -3.95793289e-01 -2.31548071e-01 -1.01895738e+00 -9.51057374e-01
1.24795806e+00 -3.84625569e-02 -1.71738017e+00 -8.01326871e-01
5.35531521e-01 -6.41367882e-02]
[ -9.73205209e-01 -2.42792264e-01 1.36213648e+00 1.56002438e+00
-1.11646473e+00 -3.59910250e-01 9.14123178e-01 9.75088835e-01
-1.16207206e+00 -6.87340617e-01]
[ 4.90442544e-01 -1.87386227e+00 -7.08038151e-01 -6.59183800e-01
8.03469121e-02 -1.24559033e+00 -2.93890119e-01 -2.18965203e-01
-1.08279467e+00 -1.75346043e-02]
[ -5.60817599e-01 1.08259249e+00 1.66278481e+00 -3.39774370e-01
4.28758174e-01 5.59271693e-01 7.63876081e-01 3.77926648e-01
8.50065351e-01 1.05124724e+00]
[ 1.75331545e+00 -6.33312404e-01 -1.00467913e-01 -1.78025097e-01
-1.31002116e+00 1.90098214e+00 8.45698237e-01 -1.42502522e+00
-6.73001707e-01 6.89108729e-01]
[ -1.73850000e+00 -9.80621397e-01 -3.26363951e-01 -5.00204444e-01
-5.31045079e-01 -3.39034826e-01 -3.57518107e-01 -3.73725556e-02
-1.26822579e+00 -1.38264406e+00]]
#####################(6)#############
<dtype: 'float32_ref'>
[[ -1.33923304e+00 3.98314148e-01 -1.05487180e+00 -2.22615644e-01
7.82311618e-01 9.53226268e-01 -2.97039151e-01 -3.89685869e-01
8.23029280e-01 7.19715893e-01]
[ 1.04759359e+00 8.69891942e-01 -5.51353216e-01 -4.16979402e-01
-8.62451375e-01 -1.88378954e+00 1.63407588e+00 -1.31232488e+00
-1.96803153e+00 -4.86700237e-01]
[ -3.07712853e-01 9.84556377e-02 4.30263966e-01 1.04724443e+00
7.22615659e-01 -5.49771845e-01 -1.07801104e+00 -3.93206358e-01
7.11512685e-01 9.57030654e-01]
[ -1.05264592e+00 6.57385737e-02 7.53750354e-02 1.01429641e+00
-8.63034368e-01 1.23717473e-03 -6.88091516e-01 -3.96133095e-01
8.48116100e-01 -9.45674896e-01]
[ 5.37974119e-01 4.54147071e-01 -2.98751473e-01 -1.59583509e+00
4.50350285e-01 6.21135473e-01 -1.53476131e+00 -1.97713211e-01
8.77439082e-01 4.83142734e-01]
[ -5.70582092e-01 -5.23053765e-01 1.98891927e-02 8.01557481e-01
-3.45719725e-01 1.27735651e+00 1.71628571e+00 -7.70039737e-01
6.76081061e-01 8.73943627e-01]
[ 1.96820140e+00 -9.69326258e-01 -7.51312554e-01 -1.13384604e+00
-6.39117777e-01 -7.42796242e-01 9.72097814e-01 1.74299920e+00
7.48745322e-01 2.23225936e-01]
[ -2.75906771e-01 -1.16707611e+00 -1.25743651e+00 -7.03301072e-01
-1.98549139e+00 -7.08913743e-01 -3.58558416e-01 3.72454494e-01
-5.64896911e-02 8.41890931e-01]
[ -1.32631826e+00 8.53675187e-01 -1.28031313e-01 2.12832183e-01
-2.22371653e-01 -9.89087045e-01 2.03618892e-02 -1.93884909e+00
-1.28941548e+00 2.91048825e-01]
[ -1.33420026e+00 5.87837324e-02 -1.05547898e-01 2.05826104e-01
1.52838349e+00 1.29717004e+00 -5.17632477e-02 -1.08887863e+00
-3.42454642e-01 -1.61216035e-02]]
None
(10, 10)
###################(7)###############
[[ 1. 1.]
[ 2. 2.]]
