有效地将自定义CSV文件读入Python

Question

I am trying to learn Python and started with this task of trying to import specific csv files in a given folder into a Python Data Type and then further processing the data. I am struggling with the part where the data needs to be imported into Python . I need this to be efficient. I tried a couple of things and attempted a couple of approaches based on suggestions provided in the forums and other web pages - all of them resulting in one problem or other . If any one can help solve this , the help would be greatly appreciated.

Note: I have already imported pandas as pd

Approach 1 :

DF = pd.read_csv(FilePath)

This yields the following error :

Error tokenizing data. C error: Expected 1 fields in line 13, saw 2

Approach 2 :

DF = pd.read_csv(FilPath, skiprows=3)

This also yields the same error :

Error tokenizing data. C error: Expected 1 fields in line 13, saw 2

Approach 3 :

data = pd.read_csv(FilePath, error_bad_lines=False)

This skips every row and reads one character per line . This makes processing the data any further hard.

https://dl.dropboxusercontent.com/u/32778128/Test.csv

If any one has any suggestions to fix this problem , I would greatly appreciate the help.

Best U

Answer 1

When you say "DataFrame" what you should be using is the Pandas library. Pandas gives you the Pandas DataFrame where you can easily manipulate import csv files and start manipulating the data. You should look into the pandas.read_csv function specifically. It will do what you're asking and more. Look into the "skiprows" argument if you need to filter by row.

You can get a DataFrame object by doing the following:

import pandas

df = pandas.read_csv('boing.csv')  # Creates dataframe from specified CSV file

If you need more than that then you can refer to the documentation linked above, pandas.read_csv takes too many arguments to list here. I hope this helps.

Example with skiprows :

df = pandas.read_csv('boing.csv', skiprows=2)

This will give you a DataFrame skipping the first two rows of your CSV file. You can change 2 to any number of headers you have. When using skiprows , make sure the you are not skipping a row that is representative of the actual data in the file.