未名方略

系统学习tensor

计算范数常用

具有求导数的函数

函数包

https://tensorflow.google.cntensorflow2.5文单

argmax of logits and probabilities are same

logit和概率的区别

可以仅仅训练不验证

get rid of tensorflow 1.X batch_norm bug

没什么是重装解决不了的

服务器版操作系统

Dice coefficient

conda create --name tensorflow python=3.7conda update -n base -c defaults conda升级spydersudo apt install spyder3或source activate tensorflowconda install spydersource activate tensorfl...

With probability keep_prob, outputs the input element scaled up by 1 / keep_prob, otherwise outputs 0. The scaling is so that the expected sum is unchanged.dropout操作目的是降低过拟合的发生率，其原理受到人脑神经网络启发。那么...

IOU = true_positive / (true_positive + false_positive + false_negative)Calculate per-step mean Intersection-Over-Uniontf.contrib.metrics.streaming_mean_ioutf.metrics.mean_iou方法1im...

重装系统蔑视

L1 regularization encourages sparsity.L2 penalizes the optimization steps.

generator，字面翻译即引擎。可以源源不断批量生成数据。看看keras经典例子def generate_arrays_from_file(path): while True: with open(path) as f: for line in f: # create numpy arrays of in...

例1import tensorflow as tfimport numpy as npBATCH_SIZE = 2train_x = np.array([1.0, 2.0, 3.0, 4.0, 5.0])train_y = np.sin(train_x)ds = tf.data.Dataset.from_tensor_slices((train_x, train_y...

tf.tile( input, multiples, name=None)Args:input: ATensor. 1-D or higher. multiples: ATensor. Must be one of the following types:int32,int64. 1-D. Length must be the same as the ...

tf.nn.rnn_cell.BasicLSTMCell__init__( num_units, forget_bias=1.0, state_is_tuple=True, activation=None, reuse=None, name=None, dtype=None, **kwargs)Args:num_uni...

以VGG net为例import osimport syssys.path.append("/***/models-master/research/slim")from nets import vgg 

Create an op that groups multiple operations.When this op finishes, all ops in inputs have finished. This op has no output.有点像批处理，法术定序。import tensorflow as tfa = tf.constant([5.0])b = tf.con...

y_pred = tf.clip_by_value(y_pred, 10e-8, 1.-10e-8)后接log防止出现log0。import tensorflow as tfv=tf.constant(([1.0,2.0,3.0],[4.0,5.0,6.0]))with tf.Session() as sess: print(tf.clip_by_value(v,2.5,...

Stacks a list of rank-Rtensors into one rank-(R+1)tensor.import tensorflow as tfa = tf.constant([[1,2,3],[4,5,6]]) b = tf.constant([[7,8,9],[10,11,12]])ab1 = tf.stack([a,b],axis=0)ab2 = tf.st...

import tensorflow as tfdef MSE_cost(out,Y): cost = tf.reduce_mean(tf.square(out-Y)) return costdef multiclass_cost(out,Y): cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logit...

除了知道返回是坐标，还须记得：tf.argmax(array, 1) 指的是 axis=1tf.argmax(array)默认axis=0 import numpy as npimport tensorflow as tf c = np.array([[1,8,3],[4,5,6]]) d = np.array([[-1,2,0],[7,8,-9]]) f1 = t...

该命令一般用于cost functionimport tensorflow as tfN_CLASSES = 5labels = [1,4,0,2,3,0,1,1]with tf.Session() as sess: sess.run(tf.global_variables_initializer()) a = sess.run(tf.one_hot(label...

目录 get_model_varibales()和get_variables() tf.valuable_scope() tf.name_scope()名称覆盖机制 slim升级至tf.contrib.framework.arg_scope()/tf.contrib.layers get_model_varibales()和get_variables()...

In mathematics, a fractal is a detailed, recursive, and infinitely self-similar mathematical set. import tensorflow as tfimport numpy as np# 导入可视化库import PIL.Imagefrom cStringIO import Strin...

numpy.ndarray.max()   和 numpy.amax() 一样。numpy,amax()Return the maximum of an array or maximum along an axis.关于axis, 请看https://blog.csdn.net/Arctic_Beacon/article/details/83307785numpy.maximum(...

sess.run()这行命令前面，加上如下：sess = tf.Session(config=tf.ConfigProto(allow_soft_placement=True, log_device_placement=True)) 终端的device mappingDevice mapping:/job:localhost/replica:0/task:0/device:GP...

全连接实现其实就是矩阵乘

首先，牢记几个函数，tf.equal()，tf.argmax(array, 1)，tf.reduce_mean()import tensorflow as tf import numpy as np A = [1,3,4,5,6]B = [1,3,4,3,2] with tf.Session() as sess: print(sess.run(tf.equal(A, B...

import tensorflow as tfimport tensorflow.contrib.slim as slim# Model Variablesweights = slim.model_variable('weights', shape=[5, 5, 3 , 3], ...

import tensorflow as tfimport numpy as npparams = np.arange(23)a = tf.gather(params,1)with tf.Session() as sess: print(sess.run(a)) params = np.arange(24).reshape(2,-1)a = tf.gather(pa...

目录tf.argmax() tf.reduce_mean()tf.argmax(), 自带降低一维import tensorflow as tfimport numpy as npx = np.ones((2,3,3,1))z = tf.argmax(x,axis = 3)print(z.get_shape())with tf.Session() as sess: ...

import tensorflow as tftf.reset_default_graph()name = 'a1'x = tf.Variable(tf.random_normal(shape=[5,5], mean=0, stddev=1), name='a2') x = tf.reshape(x,[1,5,5,1])with tf.name_scope(name): ...

例1>>>p = lambda x,y:x+y>>>p(3,4)7例2import tensorflow as tftf.reset_default_graph()x = tf.constant(10)y = tf.constant(5)a=2b=3 z = tf.multiply(a, b)result = tf.cond...

from keras.utils import np_utilsN_CLASSES = 3label = [0,0,0,1,1,1,2,2,2]train_label = np_utils.to_categorical(label, N_CLASSES)train_labelOut[21]:array([[1., 0., 0.],       [1., 0., 0.],    ...