pytorch随机数种子设置 if args.seed is not None: torch.manual_seed(args.random_seed) torch.cuda.manual_seed(args.random_seed) torch.cuda.manual_seed_all(args.random_seed) np.random.seed(args.random_seed) random.seed(args.random_seed) tor
latex自定义图片命名前缀以及设定表格名称范围与大小 自定义前缀\NeedsTeXFormat{LaTeX2e}[2007/10/19]\ProvidesClass{cpbtex}[2014/11/12 v1.0 LaTeX Template for Chinese Physics B]\LoadClass[a4paper]{article}[2007/10/19]\RequirePackage{amsmath,amsfonts,amssymb,bm,CJK,ccmap}\RequirePackage{graphics,epstopdf,ifpd
论文绘图与合成图片过程中常见问题 1、PIL格式与cv2格式的转换问题详见Stack overflowTo convert from PIL image to OpenCV use:import cv2import numpy as npfrom PIL import Imagepil_image=Image.open("demo2.jpg") # open image using PIL# use numpy to convert the pil_image into a numpy arraynumpy_image=
matplotlib相关注意事项 1、正常显示中文与正负号plt.rcParams['font.sans-serif']=['SimHei'] # 正常显示中文plt.rcParams['axes.unicode_minus']=False # 正常显示负号2、运行过程中不显示绘图import matplotlibmatplotlib.use('Agg') # 不显示绘图import matplotlib.pyplot as plt这里需要注意，matplotlib.use('Agg')需要在导入pyplot前
pytorch中几个常见的需要重写的类 文章目录1 Dataset类2 自定义transform3 自定义网络结构由于是初学者，目前在学习过程中遇到的就以上三类，如果还有其他的话欢迎补充。1 Dataset类有时候需要使用我们自己的数据集，而非pytroch自带的数据集，这时候就有必要改写Dataset类了。相关操作见 Dataset类的改写，里面写得比较详细了，摘录几句比较重要的话。当我们集成了一个 Dataset类之后，我们需要重写 __len__ 方法，该方法提供了dataset的大小； __getitem__ 方法， 该方法支持
解决tensorflow_gpu-2.3.1 出现“Could not load dynamic library ‘cudart64_101.dll‘的问题 先上问题import tensorflow as tf2020-11-30 09:54:16.112235: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found2020-11-30 09:54:16.112420: I tensorflow/stream_
Python语言Codewars实战——Sudoku Solution Validator(4 kyu) Description:Sudoku BackgroundSudoku is a game played on a 9x9 grid. The goal of the game is to fill all cells of the grid with digits from 1 to 9, so that each column, each row, and each of the nine 3x3 sub-grids (also known as blocks) contain all of the
Python实战——VAE的理论详解及Pytorch实现 2.1 设定优化目标这一部分引入是因为一个问题：在通过采样的方法(蒙特卡洛法)计算下面这个式子的时候可不可以走捷径(shortcut)？实际上(in practice)，对大多数的潜变量zzz而言，P(X∣z)P(X|z)P(X∣z)都是nearly zero的，因此它们对我们估计P(X)P(X)P(X)是没啥用的。这里就涉及到了VAE的核心思想了：尽量只采样那些对生成XXX 有贡献的zzz，然后用它们估计P(X)P(X)P(X) 。那么问题来了，从哪个分布里面去采样才能达到这个目的呢？不妨假设
Error loading “D:\Coding\Anaconda\lib\site-packages\torch\lib\asmjit.dll“ OSError: [WinError 126] The specified module could not be found. Error loading "C:\Users\chunc\anaconda3\lib\site-packages\torch\lib\asmjit.dll" or one of its dependencies.这个问题简直坑人，我在CSDN等国内的博客网站找的解决办法都是更新numpy，可谁知numpy没更新好，我的Spyder倒是莫名其妙地卸了，弄的我只能重装Anaco
MATLAB实战——微分方程组的解法之欧拉法与4阶龙格库塔法 实现：%%Question 1 part(b)clear all;clc;t0 = 0;x0 = 1/2;dt = 0.1;tf = 1;t_range = t0:dt:tf;x_EU = zeros(1,length(t_range));x_EU(1)= x0;x_RK = zeros(1,length(t_range));x_RK(1)= x0;for k = 1:length(t_range) - 1 x_EU(k+1) = euler_scheme(x_EU...
机器学习小知识——神经网络的L1和L2正则化 有时候在求解规划问题时，我们想要让得到的解满足一定的条件，换句话说，如果得到的解不能满足一定的条件我们更加倾向于拒绝它，而且这种倾向还是比较大的。这种情况在实际应用中比较常见，比如在准备出去旅行时，我们肯定不太希望背包里面放置特别重的东西，这时候就可以加一个条件，如果物品的重量大于一个给定的临界值，那么它的价值将大大降低，而这种大大降低可以通过给其添加一个很大的系数实现。同样的，在求解神经网络损失函数的最小值时，我们希望自变量权重向量www满足一定的条件，如果不满足，就给其狠狠地惩罚一下。这时候我们就会在
R语言Codewars实战——Sum by Factors(4kyu) Description:Given an array of positive or negative integersI= [i1,..,in]you have to produce a sorted array P of the form[ [p, sum of all ij of I for which p is a prime factor (p positive) of ij] ...]P will be sorted by increasing order of the prime nu
R语言Codewars实战——Best travel(5kyu) Description:John and Mary want to travel between a few towns A, B, C … Mary has on a sheet of paper a list of distances between these towns. ls = [50, 55, 57, 58, 60]. John is tired of driving and he says to Mary that he doesn’t want to drive more than t
R语言Codewars实战——Exponentials as fractions(4kyu) Description:The aim is to calculate exponential(x) (written exp(x)in most math libraries) as an irreducible fraction, the numerator of this fraction having a given number of digits.We call this function expand, it takes two parameters, x of which we want
R语言Codewars实战——Fibo akin(5kyu) Description:Be u(n) a sequence beginning with:u[1] = 1, u[2] = 1, u[3] = 2, u[4] = 3, u[5] = 3, u[6] = 4,u[7] = 5, u[8] = 5, u[9] = 6, u[10] = 6, u[11] = 6, u[12] = 8,u[13] = 8, u[14] = 8, u[15] = 10, u[16] = 9, u[17] = 10, u[18]
R语言Codewars实战——Twice linear(4kyu) Consider a sequence u where u is defined as follows:The number u(0) = 1 is the first one in u.For each x in u, then y = 2 * x + 1 and z = 3 * x + 1 must be in u too.There are no other numbers in u.Ex: u = [1, 3, 4, 7, 9, 10, 13, 15, 19, 21, 22, 27, …]
R语言Codewars实战——Number of trailing zeros of N!(5kyu) Description:Write a program that will calculate the number of trailing zeros in a factorial of a given number.N! = 1 * 2 * 3 * … * NBe careful 1000! has 2568 digits…For more info, see: http://mathworld.wolfram.com/Factorial.htmlExampleszeros(6) = 1#
R语言Codewars实战——Rainfall 说明：这道题是真难到我了，主要是我不清楚怎么把字母之间的数字匹配出来，我初始的时候想的是正则表达式，但是R语言的正则表达式与我在python中学到的不同，且更加麻烦复杂，再者我又对正则表达式一知半解的，脑子真就慢慢变得迷糊了，直接看了别人的解法(整半天经验没变化，难受啊)。Description:data and data1 are two strings with rainfall records of a few cities for months from January to December.
R语言Codewars实战——Last digit of a large number(5kyu) Description:Define a function that takes in two non-negative integers a and b and returns the last decimal digit of a^b. Note that a and b may be very large!For example, the last decimal digit of 9^7 is 9, since 9^7 = 4782969. The last decimal digit of (
关注
