kaggle商品数据集_挖掘Kaggle数据集·Instacart订单分析(一)

Instacart是一款在线订购日用商品的app，数据集提供了约3百万条订单记录，这里分两部分做一下简单的分析

第一部分：描述统计

第二部分：关联分析(Market-Basket)肖月：挖掘Kaggle数据集·Instacart订单分析(二)​zhuanlan.zhihu.com

第一部分：描述统计

先预览梳理下数据，做一些基本的描述性统计，画图比较下影响商品销量和回购的几个因素

import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

import seaborn as sns

import os,sys

from itertools import combinations, groupby

from collections import Counter

color = sns.color_palette()

数据集内容：

这是一套描述不同时段顾客购买行为的关系数据集，先把数据集导入进来看下基本信息，进行数据清洗。

from subprocess import check_output

print(check_output(["ls","../market_sells_orders/input"]).decode("utf8"))

df_order_products = pd.read_csv('../market_sells_orders/input/order_products__prior.csv')

print('order_products contains %s orders with columns:'%len(df_order_products))

print(' '+', '.join(df_order_products.columns.values))

df_orders = pd.read_csv('../market_sells_orders/input/orders.csv')

df_orders = df_orders[df_orders['eval_set']=='prior']

df_orders.drop(columns=['eval_set'],inplace=True)

print('orders contains %s orders with columns:'%len(df_orders))

print(' '+', '.join(df_orders.columns.values))

df_aisles = pd.read_csv('../market_sells_orders/input/aisles.csv')

print('aisles contains %s aisles with columns:'%len(df_aisles))

print(' '+', '.join(df_aisles.columns.values))

df_department = pd.read_csv('../market_sells_orders/input/departments.csv')

print('department contains %s departments with columns:'%len(df_department))

print(' '+', '.join(df_department.columns.values))

df_products = pd.read_csv('../market_sells_orders/input/products.csv')

print('products contains %s products with columns:'%len(df_products))

print('