PYTandFA的博客

以前写的GAN神经网络，但是训练过程并不收敛。放在这里，慢慢改！#_*_encoding:UTF-8_*_import tensorflow as tfimport numpy as npimport scipy.miscfrom glob import globimport PILsample_size = 100flags = tf.app.flagsflags.DEFINE_integer("batch_size", 100, "The number of sample in

翻到以前用c手撸的混沌粒子群算法#include <stdio.h>#include <time.h>#include <stdlib.h>#include <math.h>#include <string.h>#define N 1000#define D 2#define MAXV 0.1#define MINV -0.1#define MAXP 10#define MINP -10#define ITERATI.

手撸的word2vec，后续继续改bug。#coding+UTF-8import numpy as npfrom collections import CounterembeddingDim = 100vocabSize = 0syn0 = Nonesyn1 = Nonesyn1Neg = Nonec = 4window = np.array([x for x in range(2*c + 1) if x != c])-ctrainFile = []L = 0alpha = .

手撸的第一版Xgboost，先放上来，慢慢写bug！#coding=UTF-8import numpy as npfrom collections import Countervar_std = .5def gini(fea, val, data): """ 计算离散特征列的gini值 """ def __cal__(cls): if cls.shape[0] == 0: return .0 count...

1、KD树的构造（ongoing）#DBSCAN inspects abnormal sampleimport numpy as npfrom heapq import heappush, heappop, nsmallest, heappushpopfrom scipy.spatial import KDTreeimport matplotlib.pyplot as pltdecisionNode = dict(boxstyle="sawtooth", fc="0.8")leafNod.

利用矩阵，梯度下降算法实现的因子分解机import numpy as npLEARNING_RATE = 0.95REGULARIZATION_COEFFICIENT = 0.000001STEP_THRESHOLD = 300K = 6def sigmoid(x): return 1/(1+np.exp(-x))class factorizationMachine: #x: a sample in train_set #cross_v: it is a t...

1、概述 通过递归方式构造决策树。构造了decisionTreeNode、decisionTree两个类。ID3.py├── decisionTreeNode│ ├── __init__()│ │ ├──fea_list│ │ ├──dataset│ │ ├──split_fea│ │ ├──first_descendant│ │ ├──brother│ │ ├──cls│ │ └──split_fea_va...

自己实现的红黑树的创建与可视化，记录在下面，以后可以过来更新。 虽然红黑树的创建部分还是有些bug，创建的二叉树并不满足红黑树的4大规则，可能对红黑树算法的理解还有偏差。但是其他功能模块都运行正常。代码包括以下功能模块：1、创建二叉搜索树；2、调整二叉搜索树为红黑树；3、更新二叉树中每个节点的深度；4、更新<=节点值节点数量；5、计算二叉树的高度；...

代码#include <stdio.h>#include <stdlib.h>typedef struct Node { float value; struct Node *left; struct Node *right;}node;//查找待插入点的父节点node * findInsertLoc(node *p1, float...

快排算法输入：A[0,...,n-1]partition(A, p, q) //A待排序的数组，p为某次划分的起始index，q为某次划分的结束index i<-p pivot <- A[p] for j <-p+1 to q do if A[j] < pivot theni <- i + 1...