eerywh的博客

import tqdmimport pandas as pdimport numpy as npimport pprintdf = pd.DataFrame(np.arange(10).reshape(2,5))pprint.pprint(df) 0 1 2 3 40 0 1 2 3 41 5 6 7 8 9from sklearn.feature_extraction.text import TfidfVectorizerfrom nltk.s

第四章：特征构建：生成新的特征import pandas as pdfrom sklearn.base import TransformerMixinimport pandas as pdimport numpy as npfrom matplotlib import pyplot as pltX = pd.DataFrame({'city':['tokyo', None, 'london', 'seattle', 'san francisco', 'to

tips= sns.load_dataset("tips")tips.shapedata = tips.iloc[[2,22,222,10,30,50,70,80,201,140]]print(data) total_bill tip sex smoker day time size2 21.01 3.50 Male No...

特征理解import pandas as pdimport numpy as npimport matplotlib.pyplot as plt import seaborn as sns %matplotlib inline plt.style.use('fivethirtyeight')salary_ranges = pd.read_csv('Salary_Ranges_by_...

第一章主要是对于特征工程的介绍，阐述了特征工程的重要性。在机器学习工程师和数据科学家日常工作中，超过50%的时间事实上都是在数据的准备阶段。这其中包括存储，清洗，组织数据等等。特征工程就是这样的一个过程：将数据转换为能更好地表示潜在问题的特征，从而提高机器学习的学习性能。一个完整的特征工程包括对于数据中特征的理解，特征增强：清洗数据，特征选择，特征构建，特征转换，特征学习。最后本章还介绍了特征工程...

Data CleansingCleaning Rows with NaNsimport pandas as pddf = pd.read_csv('NaNDataset.csv')df.isnull().sum()A 0B 2C 0dtype: int64print(df) A B C0 1 2.0 31 4 ...

第三章 特征增强：清洗数据import os os.listdir()['.config', 'sample_data']!git clone https://github.com/********/Feature-Engineering-Made-Easy.gitCloning into 'Feature-Engineering-Made-Easy'...remote: Enumerating objects: 63, done.remote: Total 63 (delta 0

第二章 数据等级总结import os os.listdir()['.config', 'sample_data']!git clone https://github.com/Childish1jin/Feature-Engineering-Made-Easy.gitCloning into 'Feature-Engineering-Made-Easy'...remote: Enumerating objects: 63, done.remote: Total 63 (delta