- 博客(164)
- 资源 (3)
- 收藏
- 关注
RSS订阅原创 slake数据集数据预处理
* coding: utf-8 *“”"@Time : 2021/9/23 9:29@Author : smile 笑@File : datasets_text.py@desc :“”"import picklefrom torch.utils.data import DataLoader, Datasetfrom word_sequence import Word2Sequence, SaveWord2Vecimport torchvision.transforms as tfsim
2021-10-24 14:45:03
467
原创 实现word2sequence
coding=utf-8“”"author:leifunction: 实现的是构建词典，实现方法把句子转换为数字序列并将其翻转“”"class Word2Sequence(object):UNK_TAG = “UNK”PAD_TAG = “PAD”UNK = 0PAD = 1def __init__(self): self.dict = { self.UNK_TAG: self.UNK, self.PAD_TAG: self.PAD }
2021-08-05 10:18:17
597
原创 pytorch实现手写数字识别
coding=utf-8“”"author:leifunction: 使用pytorch完成手写数字的识别“”"import numpy as npfrom torch.utils.data import DataLoaderfrom torchvision.datasets import MNISTfrom torchvision.transforms import Compose, ToTensor, Normalizeimport torch.nn as nnimport torc
2021-08-05 10:16:40
556
原创 pytorch自实现线性回归和反向传播并画图
coding=utf-8“”"author:leifunction:“”"import torchimport matplotlib.pyplot as pltlearning_rate = 0.011、准备数据y = 3x + 0.8x = torch.rand((500, 1))y_true = x*0.3 + 0.82、通过模型计算y_predictw = torch.rand((1, 1), requires_grad=True)b = torch.empty((1, 1
2021-08-05 10:15:41
310
原创 numpy的使用
1、数据的属性¶import numpy as np创建ndarrayscore = np.array([[80, 89, 86, 67, 79],[78, 97, 89, 67, 81],[90, 94, 78, 67, 74],[91, 91, 90, 67, 69],[76, 87, 75, 67, 86],[70, 79, 84, 67, 84],[94, 92, 93, 67, 64],[86, 85, 83, 67, 80]])scorearray([[80,
2021-05-11 21:05:45
488
原创 svm支持向量机实现mnist数据集的训练和测试(准确率可达98%)
coding=utf-8“”"author:leifunction: 对mnist数据集进行模型训练，预测“”"from sklearn.svm import SVCfrom sklearn.decomposition import PCAfrom sklearn.model_selection import train_test_splitimport pandas as pdimport joblibimport matplotlib.pyplot as pltimport ran
2021-04-23 16:28:58
2691
原创 手动实现逻辑回归的整个过程及原理
from sklearn.datasets import load_iris, make_classificationfrom sklearn.model_selection import train_test_splitimport tensorflow as tfimport numpy as npX, Y = make_classification(n_samples=500, n_features=5, n_classes=2)x_train, x_test, y_train, y_tes
2021-04-23 15:49:24
519
原创 支持向量机训练mnist数据集并得到最佳参数
coding=utf-8“”"author:leifunction: 使用svm识别mnist“”"import timeimport numpy as npimport pandas as pdimport matplotlib.pyplot as pltfrom sklearn import svmfrom sklearn.model_selection import train_test_splitfrom sklearn.decomposition import PCAdef
2021-04-21 21:03:23
789
2
原创 特征降维和聚类的使用
coding=utf-8“”"author:leifunction:“”"from sklearn.cluster import KMeansimport pandas as pdfrom sklearn.decomposition import PCAfrom sklearn.metrics import silhouette_scoreorder_product = pd.read_csv("./data/instacart/order_products__prior.csv")pr
2021-04-19 17:59:35
234
原创 matplotlib百分百解决字体显示异常
from pylab import mpl设置显示中文字体mpl.rcParams[“font.sans-serif”] = [“SimHei”]设置正常显示符号mpl.rcParams[“axes.unicode_minus”] = False如果没有pylab，记得先pip install pylab 哦！
2021-04-13 21:19:42
228
原创 matplotlib多个坐标系显示图像
from pylab import mpl设置显示中文字体mpl.rcParams[“font.sans-serif”] = [“SimHei”]设置正常显示符号mpl.rcParams[“axes.unicode_minus”] = Falsex = range(60)y_beijing = [random.uniform(10, 15) for i in x]y_shanghai = [random.uniform(15, 25) for i in x]创建1行两列的画布fig, ax
2021-04-13 21:17:49
283
1
原创 matplotlib画图的简单使用
import randomfrom pylab import mpl设置显示中文字体mpl.rcParams[“font.sans-serif”] = [“SimHei”]设置正常显示符号mpl.rcParams[“axes.unicode_minus”] = False0.生成数据x = range(60)y_beijing = [random.uniform(10, 15) for i in x]y_shanghai = [random.uniform(15, 25) for i in
2021-04-13 21:16:46
166
原创 mongoDB在python中的使用（pymongo的简单操作）
coding=utf-8“”"author:leifunction:“”"from pymongo import MongoClient创建数据库链接对象client = MongoClient()client = MongoClient(“172.16.123.223”, 27017)print(client)权限认证，选择一个数据库。如果没有开权限认证，则不需要进行权限认真db = client[“admin”]账号密码进行验证db.authenticate(“python”,
2021-04-07 18:12:21
185
原创 python爬虫，嗅事百科（段子爬取）
coding=utf-8“”"author:leifunction:“”"import requestsfrom lxml import etreeimport jsonclass QiuShi(object):def __init__(self): self.url = "https://www.qiushibaike.com/text/" self.headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64
2021-04-07 18:09:17
580
7
原创 python爬虫，58汽车信息爬取（selenium）
coding=utf-8“”"author:leifunction:“”"import timefrom selenium import webdriverimport jsonclass Car(object):def init(self):self.url = “https://sanhe.58.com/ershouche/”options = webdriver.ChromeOptions()options.binary_location = r"D:\文件2\Google\C
2021-04-05 18:05:00
673
3
原创 python爬虫，虎牙房间爬取（selenium）
coding=utf-8“”"author:leifunction: 虎牙直播爬取“”"import timefrom selenium import webdriverclass HuYa(object):def init(self):self.url = “https://www.huya.com/l”options = webdriver.ChromeOptions()options.binary_location = r"D:\文件2\Google\Chrome\Applica
2021-04-05 18:04:11
352
原创 python爬虫，斗鱼房间爬取（selenium）
coding=utf-8“”"author:leifunction:斗鱼网站的爬取“”"import timefrom selenium import webdriverclass DouYu(object):def __init__(self): self.url = "https://www.douyu.com/directory/all" options = webdriver.ChromeOptions() options.binary_location =
2021-04-05 18:02:52
319
原创 python爬虫，qq空间登录（selenium在frame的切换）
coding=utf-8“”"author:leifunction:“”"from selenium import webdriverurl = “https://qzone.qq.com/”options = webdriver.ChromeOptions()options.binary_location = r"D:\文件2\Google\Chrome\Application\chrome.exe"driver = webdriver.Chrome(r"D:\文件\软件\chromed
2021-04-05 18:01:52
299
原创 python爬虫，人人网登录
coding=utf-8“”"author:leifunction:“”"import requestsimport js2pyimport jsondef login():# 创建session对象session = requests.session()# 设置请求头session.headers = {“User-Agent”: “Mozilla/5.0 (iPhone; CPU iPhone OS 13_2_3 like Mac OS X) AppleWebKit/605.1.
2021-04-05 18:00:29
233
原创 python爬虫，百度贴吧（巴塞罗那）
coding=utf-8“”"author:leifunction:“”"import requestsfrom lxml import etreeimport jsonclass TieBaBaSa(object):def __init__(self, name): self.url = "https://tieba.baidu.com/f?ie=utf-8&kw={}".format(name) self.headers = {"User-Agent": "Mo
2021-04-05 17:59:33
373
原创 python爬虫，百度贴吧
coding=utf-8“”"author:leifunction:“”"import requestsfrom lxml import etreeclass TieBa(object):def __init__(self, name): self.url = "https://tieba.baidu.com/f?kw={}&ie=utf-8&pn=0".format(name) self.headers = { "User-Agent": "
2021-04-05 17:58:33
162
原创 python爬虫，百度翻译
coding=utf-8“”"author:leifunction:百度翻译“”"import jsonimport requestsclass Baidu(object):def __init__(self, word): self.url = "https://fanyi.baidu.com/sug" self.headers = { "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) Appl
2021-04-05 17:57:22
204
原创 chrome和webdriver(selenium)的使用，解决selenium.common.exceptions.WebDriverException: Message: unknown ...
解决chrome+selenium使用时出现：selenium.common.exceptions.WebDriverException: Message: unknown error: cannot find Chrome binary。有时候可能是我们的selenium和chrome对应版本有问题，或者未将chrome和selenium加入到环境变量。但有时当我们这些都解决后还是未能自动打开chrome，还会报这样错误，那我们就可以用如下的方式进行修改。其地址：是对应于你们的chrome_webdri
2021-04-03 13:52:44
511
1
原创 python学生管理系统的实现（面向对象，含注释，保存）
coding=utf-8“”"author:leifunction:“”"定义学生类class Student(object):def init(self, name, gender, tel):self.name = nameself.gender = genderself.tel = teldef __str__(self): return "{}, {}, {}".format(self.name, self.gender, self.tel)定义学生管理类clas
2021-03-31 14:50:55
496
原创 岭回归（加入正则化的线性回归）的使用
coding=utf-8“”"author:leifunction: L2正则化 解决过拟合，使得w的每个元素都很小，都接近于0越小的参数说明模型越简单，越简单的模型则越不容易产生过拟合现象线性回归：容易出现过拟合，为了把训练集数据表现的更好，可使用L2正则化：Ridge：岭回归，带有正则化的线性回归，解决过拟合。回归解决过拟合“”"岭回归from sklearn.linear_model import Ridgefrom sklearn.preprocessing import Sta
2021-03-20 17:40:25
439
原创 逻辑回归之癌症预测
coding=utf-8“”"author:leifunction: 逻辑回归做二分类“”"import pandas as pdfrom sklearn.linear_model import LogisticRegressionimport numpy as npfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import StandardScalerfrom sklearn
2021-03-19 19:36:47
434
原创 sklearn中梯度下降法和正规方程在神经网络中的应用
coding=utf-8“”"author:leifunction:“”"from sklearn.datasets import load_bostonfrom sklearn.linear_model import LinearRegression, SGDRegressorfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import StandardScalerfrom sk
2021-03-19 19:35:54
201
原创 聚类中K-Means和DBscan的使用
coding=utf-8“”"author:leifunction:“”"import pandas as pdfrom sklearn.cluster import KMeansbeer = pd.read_csv(“data.txt”, sep="")X = beer[[“calories”, “sodium”, “alcohol”, “cost”]]km = KMeans(n_clusters=3).fit(X)km2 = KMeans(n_clusters=2).fit(X)b
2021-03-19 19:34:50
206
原创 K近邻和朴素贝叶斯算法的使用
coding=utf-8“”"author:leifunction: k近邻算法，欧式距离，两点之间的距离公式需要进行标准化处理“”"from sklearn.neighbors import KNeighborsClassifierfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import StandardScalerfrom sklearn.datasets import f
2021-03-19 19:33:31
209
原创 机器学习泰坦尼克号获救预测
coding=utf-8“”"author:leifunction: 决策树“”"from sklearn.tree import DecisionTreeClassifierfrom sklearn.feature_extraction import DictVectorizerfrom sklearn.model_selection import train_test_splitfrom sklearn.tree import export_graphvizfrom sklearn.d
2021-03-19 19:32:07
290
1
原创 输入一个长方形的长和宽（整数），输出该长方形的周长C和面积S，要求格式如例（请注意打印“C:”、“S:”，使用半角冒号，参考样例输出）。
1952：输入一个长方形的长和宽（整数），输出该长方形的周长C和面积S，要求格式如例（请注意打印“C:”、“S:”，使用半角冒号，参考样例输出）。#include<stdio.h>int main(){int a,b;scanf("%d %d", &a, &b);printf("C:%d\nS:%d\n", 2*(a+b), a*b);return 0;}...
2021-03-19 19:30:19
3476
原创 pytorch实现mnist数据集的训练和预测
coding=utf-8“”"author:leifunction:“”"import osimport torchfrom torch.optim import Adamfrom torch.nn import functional as Ffrom torch import nnfrom torch.utils.data import DataLoaderfrom torchvision.datasets import MNISTfrom torchvision.transfor
2021-03-19 19:30:07
671
原创 sklearn的特征选择
coding=utf-8“”"author:lei 5function: 特征选择“”"from sklearn.feature_selection import VarianceThreshold # 特征选择from sklearn.decomposition import PCAnum_list = [[0, 2, 0, 3], [0, 1, 4, 3], [0, 1, 1, 3]]def var():“”"特征选择-删除低方差特征:return:“”"var = Va
2021-03-19 19:29:24
210
原创 sklearn特征抽取及tfidf的使用
coding=utf-8“”"author:leifunction: 特征抽取“”"from sklearn.feature_extraction import DictVectorizerfrom sklearn.feature_extraction.text import CountVectorizerfrom sklearn.feature_extraction.text import TfidfVectorizerimport jiebatext_list = [“life is
2021-03-19 19:28:34
382
原创 sklearn的特征处理
coding=utf-8“”"author:leifunction:“”"from sklearn.preprocessing import MinMaxScaler # 归一化from sklearn.preprocessing import StandardScaler # 标准化from sklearn.preprocessing import Imputer # 数据预处理import randomimport numpy as npnum_list = [[random.
2021-03-19 19:26:19
181
原创 pytorch的数据加载
coding=utf-8“”"author:leifunction:“”"import torchfrom torch.utils.data import Dataset, DataLoaderimport mathdata_path = “./data/SMSSpamCollection”完成数据集类class MyDataset(Dataset):def init(self):self.lines = open(data_path, encoding=“utf8”).readli
2021-03-18 08:35:36
216
2
原创 pytorch实现线性回归
coding=utf-8“”"author:leifunction:“”"from torch.optim import SGDimport torchfrom torch import nn定义一个device对象 如果要转换为gpu上运行device = torch.device(“cuda” if torch.cuda.is_available() else “cpu”)准备数据x = torch.rand([500, 1]).to(device)y_true = 3 * x
2021-03-18 08:34:25
180
原创 卷积神经网络完成mnist数据集的训练和测试
coding=utf-8“”"author:leifunction:“”"import numpy as npimport tensorflow as tf下载并载入Mnist 手写数字库（55000 * 28 * 28）from tensorflow.examples.tutorials.mnist import input_datamnist = input_data.read_data_sets(‘mnist_data’, one_hot=True)one_hot 独热码(enco
2021-03-18 08:31:22
1336
原创 手动搭建神经网络实现mnist数据集的训练
import tensorflow as tffrom tensorflow.examples.tutorials.mnist import input_data载入数据集mnist = input_data.read_data_sets(“mnist_data”, one_hot=True)每个批次的大小batch_size = 100计算一共多少个批次n_batch = mnist.train.num_examples // batch_size定义两个placeholder 一张图片
2021-03-18 08:30:14
429
原创 tensorflow完成mnist数据集训练和验证
coding=utf-8“”"author:leifunction:“”"import tensorflow as tfimport numpy as npimport matplotlib.pyplot as pltfrom tensorflow.examples.tutorials.mnist import input_datamnist = input_data.read_data_sets("./mnist_data/", one_hot=True)trainimgs, trai
2021-03-18 08:28:17
305
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人