RogerFedereYY的博客

jupyter lab中pyecharts的.render_notebook()无法显示问题使用pyecharts自带的config来申明一下使用pyecharts的运行环境from pyecharts.globals import CurrentConfig, NotebookTypeCurrentConfig.NOTEBOOK_TYPE = NotebookType.JUPYTER_LABCurrentConfig.ONLINE_HOST另起单元格import numpy as np

运行以下代码即可# 解决中文不能显示的问题plt.rcParams['font.sans-serif'] = ['SimHei']# 中文字体无法正常显示负号.plt.rcParams['axes.unicode_minus'] = False# seabornsns.set_style('whitegrid', {'font.sans-serif':['simhei', 'Arial']})...

3d图# 方式一plt.subplot(projection='3d')# 方式二# get current axesplt.figure(figsize=(14, 10))axes = plt.gca(projection='3d')空间直线方式一x = np.linspace(0, 10, 20)y = np.linspace(2, 8, 20)z = 2*x + 5*y + 3plt.plot(x, y, z)方式二x,y,z = array([22.5, 50

导包import pandas as pdimport numpy as npimport stringfrom pandas import DataFrame, Series生成字典dic = dict(zip(string.ascii_lowercase,string.digits))dic"""{'a': '0', 'b': '1', 'c': '2', 'd': '3', 'e': '4', 'f': '5', 'g': '6', 'h': '7', 'i':

分类的目标变量是标称型数据，分类解决离散问题；回归将会对连续型的数据做出预测，回归解决连续问题一、KNN分类问题：from sklearn.neighbors import KNeighborsClassifier回归问题：from sklearn.neighbors import KNeighborsRegressor原理：K-近邻算法采用测量不同特征值之间的距离方法进行分类（k-Nearest Neighbor，KNN）优点：精度高、对异常值不敏感、无数据输入假定。缺点：时间复杂度高

import numpy as npn = np.random.randint(1,10,(4,3))n"""array([[9, 8, 8], [5, 9, 5], [8, 9, 6], [6, 3, 6]])"""n[[1,2]] = n[[2,1]]n"""array([[9, 8, 8], [8, 9, 6], [5, 9, 5], [6, 3, 6]])"""

whiten 白化. 每个特征的标准偏差是否一致默认default，一般改成whiten=True

读取seeds.tsv文件（文件已经上转到本博客下载中心，可免费下载），最后一列是小麦品种，其他列是小麦特征导包import numpy as npimport pandas as pdimport matplotlib.pyplot as plt数据获取seeds = pd.read_csv('./seeds.tsv',sep='\t',header=None)data = seeds.iloc[:,:-1]data = np.array(data)target = seeds.

导包import numpy as npimport matplotlib.pyplot as pltfrom sklearn.datasets import make_blobs数据data, target = make_blobs(centers=2)plt.scatter(data[:,0], data[:,1], c=target)逻辑斯蒂函数# 目标函数def aim(X,theta): return np.dot(X,theta.T)# 逻辑斯蒂函数def

导包import numpy as npimport matplotlib.pyplot as plt数据x = np.random.random(100)y = x*2 + 3 + np.random.normal(size=100)X = x.reshape(-1,1)plt.scatter(X,y)X0 = 1X_C = np.hstack((np.ones(X.shape),X))X_C.shape损失函数def J(b_w,X_C,y): return np.