8rep --Pandas string delete code
Delete unnecessary characters with Pandas
from sqlalchemy import create_engine
import pandas as pd
fname="{Pth}.csv"
reader = pd.read_csv(fname, chunksize=1000, sep='\t',low_memory = False)
df_all = reader.get_chunk() #chunk to dataframe
#Row unit missing value judgment
df_all.isnull().any(axis=1)
#Removed some NaN lines
df_all=df_all.dropna(how="any")
#Specify delete character
replace_str={
'@': '',
'\+': '',
'\$': '',
'\|': '',
'\<': '',
'\>': '',
'\-': '',
'\;': ''
}
#Delete a specific string
df_all["comment"]=df_all["comment"].replace(replace_str, regex=True, encoding='utf-8')
print(df_all)
df_all.to_csv('{Pth}/replace.csv', sep='\t', index=False)