luoganttcc的博客

link

Am×n×..×km \times n \times ..\times km×n×..×k 和A1×k1\times k1×k 矩阵相加相当于在最后一个维度import numpy as np a=np.array(range(3*4)).reshape([3,4])b=np.array([0.2]*4)print('a.shape=',a.shape)print('b.shape=',b.shape)print(a+b)print('-'*20+'我是分割线'+'-'*20)

对应位置的索引赋值这里是一维坐标赋值import tensorflow as tftensor = [0, 0, 0, 0, 0, 0, 0, 0] # tf.rank(tensor) == 1indices =[ [1], [3],[4], [7]] # num_updates == 4, index_depth == 1updates = [ 9, 10, 11, 12] # num_updates == 4print(tf.

tensorflow 动态数组随时可以读取import tensorflow as tfta = tf.TensorArray(tf.float32, size=0, dynamic_size=True, clear_after_read=False)ta = ta.write(0, 10)ta = ta.write(1, 20)ta = ta.write(2, 30)print(ta.read(0))print(ta.read(1))print(ta.read(2))prin

tf.lookup.StaticHashTable 本质是tensorflow 内置字典，在yolov3 tf代码中多次应用def load_tfrecord_dataset(file_pattern, class_file, size=416): LINE_NUMBER = -1 # TODO: use tf.lookup.TextFileIndex.LINE_NUMBER class_table = tf.lookup.StaticHashTable(tf.lookup.TextF

tf.data.DatasetAPI支持写入的描述性和高效的输入管线。Dataset用法遵循一个常见模式：从输入数据创建源数据集。应用数据集转换来预处理数据。迭代数据集并处理元素。迭代以流式方式发生，因此不需要将完整数据集放入内存中。import tensorflow as tfdataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])for element in dataset: print(element)dataset =

tensorflow-TFRecord 文件详解TFRecord 是 tensorflow 内置的文件格式，它是一种二进制文件，具有以下优点：统一各种输入文件的操作更好的利用内存，方便复制和移动将二进制数据和标签(label)存储在同一个文件中import os import numpy as npimport tempfileimport tensorflow as tf# example_path = os.path.join(tempfile.gettempdir(

import numpy as npfrom tensorflow.keras.layers import Lambdaimport tensorflow as tfx=np.array([[1,2],[3,4]])x = Lambda(lambda x: x**2)(x)print(x)print('---------------------------------------')x = Lambda(lambda x: tf.reshape(x, (-1,x.shape[1],

import tensorflow as tfx= tf.constant( [[2,3],[4,5],[6,7]], tf.int32)print(x.numpy())[[2 3] [4 5] [6 7]]x1=tf.reshape(x, shape = (tf.shape(x)[1], tf.shape(x)[0]))print(x1.numpy())[[2 3 4] [5 6 7]]

import tensorflow as tfx= tf.constant([ [[2,3],[4,5]], [[6,7],[8,9]] ], tf.int32)d1=tf.range(0, tf.shape(x)[0])print(d1.numpy())d2=tf.tile( d1, multiples=[tf.shape(

不知道大家用没用过CAD ，CAD中有一个方法叫阵列，tf.title 本意就是阵列import tensorflow as tfa = tf.constant([[1,2,3],[4,5,6]], tf.int32)b = tf.constant([2,3], tf.int32)tf.tile(a, b)<tf.Tensor: shape=(4, 9), dtype=int32, numpy=array([[1, 2, 3, 1, 2, 3, 1, 2, 3], [4.

tf.ragged.map_flat_values 可以理解不规则矩阵的mapimport tensorflow as tf rt = tf.ragged.constant([[1, 2, 3], [], [4, 5], [6]])tf.ragged.map_flat_values(tf.ones_like, rt).to_list()tf.ragged.map_flat_values(tf.multiply, rt, rt).to_list()tf.ragged.map_flat_val

git链接参考链接训练模型#!/usr/bin/env python3# -*- coding: utf-8 -*-"""Created on Sat Mar 16 22:26:43 2019@author: lg"""#coding=utf-8 # 载入MINIST数据需要的库from tensorflow.examples.tutorials.mnist import...

#!/usr/bin/env python3# -*- coding: utf-8 -*-"""Created on Thu Oct 25 13:41:35 2018@author: lg"""import numpy as npimport pandas as pdimport tensorflow as tfimport tensorflow.contrib.rnn as...

#!/usr/bin/env python3# -*- coding: utf-8 -*-&quot;&quot;&quot;Created on Thu Oct 25 15:29:59 2018@author: lg&quot;&quot;&quot;import tensorflow as tfimport numpy as npimport matplotlib.pyplot as pltmoney=np.ar

网上看到很多tf.pad的文章，很多都语焉不详，需自己动手操作才能真正理解，pad,一言以蔽之，就是在矩阵的周围补0。对于二维矩阵，就是在 上 ，下，左，右补0import numpy as npimport tensorflow as tf # 创建一个二维变量,默认执行CONSTANT填充vct = tf.Variable(tf.ones([3, 4]), name=&amp;quot;vct&amp;quot;...

在登录远程服务器的时候使用命令：ssh -L 16006:127.0.0.1:6006 name@server_ipnohup python file_name.py &amp;gt;p1.log &amp;gt;tensorboard --logdir=./log在本机浏览器输入：http://127.0.0.1:16006/...

#!/usr/bin/env python3# -*- coding: utf-8 -*-&quot;&quot;&quot;Created on Mon Oct 22 13:04:34 2018@author: lg&quot;&quot;&quot;#!/usr/bin/env python3# -*- coding: utf-8 -*-import tensorflow as tfimport matp

import tensorflow as tfimport numpy as np## prepare the original datawith tf.name_scope('data'): x_data = np.random.rand(100).astype(np.float32) y_data = 0.3*x_data+0.1##creat parameter...

原文链接

import tensorflow as tf; import numpy as np; x = tf.constant([[1, 2, 3], [4, 5, 6]])cc=tf.transpose(x)init = tf.initialize_all_variables()sess = tf.Session()sess.run(init) # Very import...

import numpy as npimport tensorflow as tfa=np.reshape(range(24),(4,2,3))sess=tf.InteractiveSession()b= tf.split(a,1,0)c=sess.run(b)切割cOut[37]: [array([[[ 0, 1, 2], [ 3, 4, 5]]...

训练模型import tensorflow as tfimport numpy as npimport matplotlib.pyplot as pltmoney=np.array([[109],[82],[99], [72], [87], [78], [86], [84], [94], [57]]).astype(np.float32)click=np.array([[11], [8]...

import numpy as np import tensorflow as tf import tensorflow.contrib.rnn as rnn import matplotlib.pyplot as plt import matplotlib.pyplot as pltTIME_STEPS=10BATCH_SIZE=128HIDDEN_UNITS1=30HIDDE...