Python Dataframe之excelcsvpicklefeatherparquetjayhdf5 文件存储格式==》存读效率对比

Posted 2021-12-12 卖山楂啦prss

今天看到一篇文章

参考：对比不同主流存储格式（csv, feather, jay, h5, parquet, pickle）的读取效率

然后我自己也试了一下，感觉发现了“新大陆”，T_T~到现在才知道还有这些存储形式，比这excel、csv快多了。

上次实习的时候，因为不知道可以存为其他格式，把多个几十个G的dataframe处理完后存为csv，过后又要读出来

心态瞬间崩了~

搞数据都搞了好久，浪费时间~拐求

文章目录

• desk 读取CSV文件
• 一、excel 存储格式（xlsx）
• 二、csv 存储格式
• 三、pickle 存储格式
• 四、feather 存储格式
• 五、parquet 存储格式
• 六、jay 存储格式
• 七、hdf5 存储格式

多的不说，直接看效果

这个结论直接借一下参考文章的，有需要的可以详细看参考文章

还有就是，数据稍微大一点，最好不要存为 excel 格式，这真的~慢

最后最后，在借用这位大兄弟的一句话

desk 读取CSV文件

读

import time
import dask
import dask.dataframe as dd
from dask.diagnostics import ProgressBar
from numba import jit
import pandas as pd
import numpy as np
import sys
# ----------------------------------------------------------------------------
switchDict = {
    0 : 'TEST',
    1 : 'ALL'
}

# 编译数据量状态开关 0为测试（读部分数据），1为全量
status = switchDict[1]
@jit
def importData(fileName):
    if status == 'TEST':
        df = dd.read_csv(fileName, header=None, blocksize="100MB").head(17000)
    else:
        df = dd.read_csv(fileName,  blocksize="64MB").compute()
    df.index = pd.RangeIndex(start=0, stop=len(df))
    return df
  
# 读正样本
t0=time.time()
t1= time.perf_counter()
with ProgressBar():
    data = importData('train.csv')
t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 3.421277699999337
wall time: 3.421303749084473
'''
print(f"当前数据框占用内存大小：{sys.getsizeof(data)/1024/1024:.2f}M") 
data.shape

一、excel 存储格式（xlsx）

存

存为excel表，真的慢到天荒~拉胯啊

import time
t0=time.time()
t1= time.perf_counter() 

data.to_excel("data.xlsx")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)

等了十几分钟，算了~

读

import time
t0=time.time()
t1= time.perf_counter() 

data_excel = pd.read_excel("./data.xlsx")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)

二、csv 存储格式

存

import time
t0=time.time()
t1= time.perf_counter() 

data.to_csv("data.csv")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 32.49002720000135
wall time: 32.48996901512146
'''

读

import time
t0=time.time()
t1= time.perf_counter() 

data_csv = pd.read_csv("./data.csv")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 7.5742819999995845
wall time: 7.574833154678345
'''

三、pickle 存储格式

Pickle：用于序列化和反序列化Python对象结构

详细百度八~

存

import time
t0=time.time()
t1= time.perf_counter() 

data.to_pickle("data.pkl.gzip")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 1.1933384000002625
wall time: 1.1980044841766357
'''

读

import time
t0=time.time()
t1= time.perf_counter() 

data_pickle = pd.read_pickle("./data.pkl.gzip")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 1.246990000000551
wall time: 1.246736764907837
'''

四、feather 存储格式

Feather：一个快速、轻量级的存储框架

网上很多推荐这个存储格式的

再见 CSV，速度提升 150 倍！

详细百度八~

---

存

import time
t0=time.time()
t1= time.perf_counter() 

data.to_feather("data.feather")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 0.5462657999996736
wall time: 0.5466225147247314
'''

读

t0=time.time()
t1=time.perf_counter() 

data_feather = pd.read_feather("./data.feather")

t2=time.time()
t3=time.perf_counter() 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 0.6685380999997506
wall time: 0.6682815551757812
'''

五、parquet 存储格式

Parquet：Apache Hadoop的列式存储格式

详细百度八~

存

import time
t0=time.time()
t1= time.perf_counter() 

data.to_parquet("data.parquet")

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 2.874607599999763
wall time: 2.874359369277954
'''

读

t0=time.time()
t1=time.perf_counter() 

data_parquet = pd.read_parquet("./data.parquet")

t2=time.time()
t3=time.perf_counter() 

print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 0.9940449000000153
wall time: 0.9959096908569336
'''

六、jay 存储格式

安装 datatable 包

pip install -i https://pypi.doubanio.com/simple/ --trusted-host pypi.doubanio.com datatable

存

import datatable as dt
t0=time.time()
t1=time.perf_counter() 

dt.Frame(data).to_jay("data.jay")

t2=time.time()
t3=time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 6.169269200000599
wall time: 6.168536901473999
'''

读

当我查看内容时，该对象是frame

t0=time.time()
t1=time.perf_counter() 

data_jay = dt.fread("./data.jay")

t2=time.time()
t3=time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 0.03480849999959901
wall time: 0.034420013427734375
'''
data_jay.shape

七、hdf5 存储格式

存

普通格式存储

import time
t0=time.time()
t1= time.perf_counter() 

####普通格式存储：
h5 = pd.HDFStore('./data.h5','w')
h5['data'] = data
h5.close()

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 2.1860209000005852
wall time: 2.186391592025757
'''

压缩格式存储

import time
t0=time.time()
t1= time.perf_counter() 

####压缩格式存储
h5 = pd.HDFStore('./data.h5','w', complevel=4, complib='blosc')
h5['data'] = data
h5.close()

t2=time.time()
t3= time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 1.9893786000002365
wall time: 1.9896411895751953
'''

读

t0=time.time()
t1=time.perf_counter() 

data_hdf5 = pd.read_hdf('./data.h5',key='data')

t2=time.time()
t3=time.perf_counter() 
 
print("cpu time:",t3-t1)
print("wall time:",t2-t0)
'''
cpu time: 1.4497185000000172
wall time: 1.4497275352478027
'''