- 对比:使用
pandas
存储数据 VS 使用写文本 方式存储数据
import pandas as pd
import time
def pandasWrite():
t0 = time.time()
colname = [str(i) for i in range(550)]
df = pd.DataFrame(columns=colname)
for i in range(100):
df.loc[len(df)] = dict(zip(colname, range(550)))
t1 = time.time()
df.to_csv("temp.csv")
print("pandas 存储数据用时:", t1-t0)
# print(df)
def fileWrite():
t0 = time.time()
colname = [str(i) for i in range(550)]
with open("temp1.txt", 'w', encoding='utf-8') as f:
f.write('t'.join(x for x in colname))
for i in range(100):
f.write('t'.join(str(x) for x in range(550)) 'n')
t1 = time.time()
print("写文本 存储数据用时:", t1-t0)
pandasWrite()
fileWrite()
输出:
代码语言:javascript复制pandas 存储数据用时: 4.545027494430542
写文本 存储数据用时: 0.03499293327331543
写文本方式,快了 接近 130 倍
工作当中踩过的坑,浪费了大半天时间,大家注意!