DeniuHe的博客

【代码】np.clip()函数用法。

Smooth Low Rank and Sparse Matrix Recovery

SimpleImputer

Dataframe 按条件替换某列中的元素

import numpy as npimport pandas as pddf = pd.read_csv(r"D:\OCdata\ObesityDataSet_raw_and_data_sinthetic.csv")# # -------按列读取# print(df["Gender"])# print(df[["Gender","Age"]])# print(df.loc[:,"Age"])# print(df.loc[:,["Gender","Age"]])## # -------.

import numpy as npfor idx in np.arange(0, 10)[::-1]: print(idx)

import matplotlib.pyplot as pltimport numpy as npclass GD(object): def __init__(self,seed=None, precision=1.E-6): self.seed = GD.get_seed(seed) # 梯度下降算法的种子点 self.prec = precision # 梯度下降算法的精度 self.path = list() .

import numpy as npa = np.array([1,2,3,4,5,6])###The first formif a.all() > 0: print("数组中所有元素都大于0")else: print("数据中有小于等于0的元素")###The second formif all(i > 0 for i in a) > 0: print("数组中所有元素都大于0")else: print("数据中有小于等于0的元素").

import numpy as npa = np.array([1,2,3,4,5,6])###The first formif a.all() > 0: print("数组中所有元素都大于0")else: print("数据中有小于等于0的元素")###The second formif all(i > 0 for i in a) > 0: print("数组中所有元素都大于0")else: print("数据中有小于等于0的元素").

if self.intLabeled: for idx in self.intLabeled: interval = deepcopy(self.Xin[idx].inter) print("labels==",self.labels) print("interval===",interval) print("theta===",self.theta) ids_lab = [] for lab in i.

单纯形法：import numpy as npfrom scipy.optimize import minimizedef rosen(x): return sum(100.0 * (x[1:] - x[:-1] ** 2.0) ** 2.0 + (1 - x[:-1]) ** 2.0)def callback(xk): print(xk)# 初始迭代点# x0 = np.random.randn(10)*2x0 = np.ones(10)print("x0::",

import numpy as npA = np.array([[1,2,3],[4,5,6]])print("上面的减去下面的，结果整体少一行：")print(np.diff(A,axis=0))print("右边的减去左边的，结果整体少一列：")print(np.diff(A,axis=1))

import numpy as npA = np.array([[1,2,3],[4,5,6]])print("把矩阵A先横向复制两次，在整体纵向复制四次：")print(np.tile(A,(2,4)))

import numpy as npa = np.array([[1,2,3],[1,2,3]])b = np.array([[1,2,3],[1,2,3]])print(a)print(b)print(a * b)print(a.T @ b)print(a @ b.T)

拿不准，再试一试 ，刚好用到import numpy as np# def rosen(x):a = np.arange(10)print("a::",a)print("去头",a[1:])print("去尾",a[:-1])

x = np.random.randn(4, 4)print(x)A = np.where(x>0, 2, -2)print(A)结果：np.where()对列表无效a = [1,2,2,3,2,5]print(np.where(a == 2))print(np.where(a == 2)[0])结果：a = np.array(a)print(np.where(a == 2))print(np.where(a == 2)[0])...

import numpy as npn_class = 5L = np.zeros((n_class - 1, n_class - 1))print(L)L[np.tril_indices(n_class-1)] = 1.print(L)

第一次发现可以这么用，我好low

import numpy as npa = np.array([[0.5,1.5,2.5],[-0.5,0.5,1.5],[-1.5,-0.5,0.5],[-3,-2,-1]])print(a)print(a < 0)##################### a[a<0] = 0# print(a)##################### b = np.where(a>0,a,0)# print(b)#################### np.maximu.

import numpy as npa = np.array([[0.5,np.nan,2.5],[-0.5,0.5,1.5],[-1.5,-0.5,0.5],[-3,-2,-1]])print(np.isnan(a))a[np.isnan(a)] = 1.5print(a)

于是就有了序分了的一个评价指标 ：最后一句话讲的非常好！论文出处：Learning to Classify Ordinal Data: The Data Replication MethodJournal of Machine Learning Research 8 (2007) 1393-1429

只要结果不要过程！博主懒地写输入。。。不懂啥是mean rank，就没救了~import numpy as npA = [[26.44, 25.8, 25.54, 25.75, 25.32, 26.04, 25.23, 25.11, 24.45],[1.9, 1.88, 4.1, 1.93, 2.88, 2.97, 3.3, 2.96, 1.8],[27.26, 28.39, 28.06, 27.45, 27.67, 30.81, 29.23, 26.62, 26.42],[1.

缺点：大概率首选离群点改进效果：首先剔除密度较小的点，作为虚拟的已标记样本点。粗放版代码：'''author:Danieldate:2021-04-03organization: CQUPTReference: D. Wu, C. Lin, J. Huang. Active learning for regression using greeding sampling. Information Sciences, 2019, 474: 90-105.'''import x.

import xlwtimport numpy as npimport pandas as pdfrom copy import deepcopyfrom sklearn.metrics import accuracy_score, mean_absolute_errorfrom collections import OrderedDictfrom mord import LogisticATfrom sklearn.model_selection import StratifiedKFol.

代码存档，读者可以绕道，没有技术内容。import xlrdimport xlwtimport numpy as npfrom pathlib import Pathfrom collections import OrderedDictread_path_1 = Path(r"D:\AOR_experiment\pointwise\uncertainty")read_path_2 = Path(r"D:\AOR_experiment\pointwise\LogitA")read_.

使用python就是为了节约撸码的时间。import numpy as npfrom sklearn.datasets import load_irisfrom sklearn.metrics.pairwise import rbf_kernelimport pandas as pdfrom scipy.linalg import block_diaga = np.array([1,2,3,4,5])print(np.outer(a,a))###验证b = a.reshap.

啥也不是import numpy as npimport pandas as pdimport xlwtfrom pathlib import Pathfrom copy import deepcopyfrom collections import OrderedDictfrom sklearn.linear_model import LogisticRegressionfrom sklearn.model_selection import StratifiedKFoldfrom .

"""Auther:Deniu HeDate:2021-03-16"""import numpy as npfrom sklearn.datasets import load_irisfrom sklearn.metrics.pairwise import rbf_kernelimport matplotlib.pyplot as pltfrom scipy.linalg import block_diagclass GDA(): def __init__(self,X,y,g.

'''Auther: DeniuHeDate:2021-03-16'''import numpy as npfrom sklearn import datasetsfrom sklearn.metrics.pairwise import rbf_kernelfrom sklearn.decomposition import KernelPCAfrom sklearn.model_selection import train_test_splitimport matplotlib.pypl.

ord_ids = np.flipud(np.argsort(self.L))

调包偷个懒import numpy as npfrom sklearn.datasets import load_irisfrom sklearn.metrics.pairwise import rbf_kernelimport pandas as pdif __name__ == '__main__': X, y = load_iris(return_X_y=True) a = [1,2,3,4,5,6,7,8] b = [110,111] A = X[.

单纯看公式推导看懂了，是没有用的。"""Auther:Deniu HeDate:2021-03-12"""import numpy as npfrom sklearn.datasets import load_irisimport matplotlib.pyplot as pltclass LDA(): def __init__(self,X,y): self.X = X self.y = y self.N = self.X.s.

LDA线性判别分析实现有监督降维，代码将iris数据降到2维"""Auther:Deniu HeDate:2021-03-12"""import numpy as npfrom sklearn.datasets import load_irisimport matplotlib.pyplot as pltclass LDA(): def __init__(self,X,y): self.X = X self.y = y s..

import numpy as npa = np.array([[1,2,3],[1,2,3],[1,2,3],[1,2,3]])b = np.array([1,2,3,4])print(a.shape)c = b.reshape(-1,1)print(c.shape)print(a/c)print(np.sum(a,1))

``$\mathbf{x}_r<_f\mathbf{x}_s$''Tab键上面那个按钮，点两下！

'''Data: 2021-01-16author: Deniu Heat: CQUPT'''import cvxpy as cpimport xlwtimport numpy as npimport pandas as pdfrom copy import deepcopyfrom sklearn.metrics import accuracy_score, mean_absolute_errorfrom collections import OrderedDictfrom sk.

import numpy as npimport pandas as pdfrom pathlib import Patha = [[1,2,3],[4,5,6],[7,8,9]]print(np.mean(a,axis=0))p = Path("D:\OCdata")names = ["Obesity", "PowerPlant-5bin", "PowerPlant-10bin"]for name in names: path = p.joinpath(name + ".cs.

import numpy as npimport pandas as pdpath = r"C:\Users\pc\Desktop\PPT素材\CCPP\ccpp.csv"data = np.array( pd.read_csv(path,header=None))print(data.shape)y = data[:,-1]ord_idx = np.argsort(y)Bins = [1913, 1913, 1914, 1914, 1914]Binss = [1913,3826,574.

import pandas as pdimport numpy as npimport osos.getcwd()os.chdir("D:\OCdata")from sklearn.preprocessing import OneHotEncoderfrom sklearn.preprocessing import LabelEncoderdata = pd.read_csv("ObesityDataSet_raw_and_data_sinthetic.csv")# print(dat.

不能自动补全，可能是因为没有写 self.