tensorflow学习笔记
Rooper
Enjoy the present!
展开
-
tf_矩阵的广播机制
在神经网络的添加层中def add_layer(inputs, in_size, out_size,activation_function = None):Weights = tf.Variable(tf. random_normal([in_size,out_size]))biases = tf.Variable(tf.zeros([1,out_size])+ 0.1)# like list转载 2017-12-20 10:58:37 · 2521 阅读 · 0 评论 -
tf_神经网络的简单搭建
import tensorflow as tf import numpy as np def add_layer(inputs, in_size, out_size, activation_function=None): Weights = tf.Variable(tf.random_normal([in_size,out_size])) biases转载 2017-12-21 11:06:13 · 746 阅读 · 0 评论 -
tensorflow_1 小试牛刀
import module import tensorflow as tf import numpy as npcreat data # creat 100 random sequences for x_data x_data = np.random.rand(100).astype(np.float32) y_data = x_data*0.1 + 0.3翻译 2017-12-18 14:46:06 · 242 阅读 · 0 评论 -
tf_2_Session会话控制
两种常用的会话控制方法:import tensorflow as tfimport numpy as np# 简单的举一个矩阵相乘的例子matrix1 = tf.constant([[3, 3]])matrix2 = tf.constant([[2], [2]])product = tf.matmul(matrix1, matrix2) # matrix转载 2017-12-19 10:56:05 · 581 阅读 · 0 评论 -
tf_3_Variable
Variable用法的简单介绍 import tensorflow as tf state = tf.Variable(0,name='counter') # 必须定义是变量它才是变量,与python语法有点不同 # print(state.name) one = tf.constant(1) new_value = tf.add(state, one) up转载 2017-12-19 11:28:20 · 198 阅读 · 0 评论 -
tf_placehoder传入值
import tensorflow as tf # 相当于定义了两个形参,或者说定义了两个占位符,在后面是用的时候还要输入具体值 input1 = tf.placeholder(tf.float32)# most type are float32 input2 = tf.placeholder(tf.float32)# most type are float32 out转载 2017-12-19 13:26:04 · 326 阅读 · 0 评论