Python 初学者 - 使用 dtale 编码（命令顺序问题）

Question

Due to a university project where I want to work and learn python, I stumbled upon the add-on dtale, which helps me analyzing party manifesto mass data.

Long story short: I added some filters (eg I only want to show rows with an edate >= 20140914. When I run the code, the filters don't seem to be applied - could you please help me with that?

import pandas as pd
df = pd.read_csv('https://manifestoproject.wzb.eu/down/data/2020b/datasets/MPDataset_MPDS2020b.csv')
d = dtale.show(df)

# DISCLAIMER: 'df' refers to the data you passed in when calling 'dtale.show'

import pandas as pd

if isinstance(df, (pd.DatetimeIndex, pd.MultiIndex)):
    df = df.to_frame(index=False)

# remove any pre-existing indices for ease of use in the D-Tale code, but this is not required
df = df.reset_index().drop('index', axis=1, errors='ignore')
df.columns = [str(c) for c in df.columns]  # update columns to strings in case they are numbers

df.loc[:, 'edate'] = pd.Series(pd.to_datetime(df['edate'], infer_datetime_format=True), name='edate', index=df['edate'].index)
d.open_browser()

So basically, my goal is to not always have to start filtering for dates etc, but that all my progress is saved and applied when running the code.

Thanks for your help!

Answer 1

There are some other arguments you can pass to pd.read_csv() that will probably help you out here:

• parse_dates : Give this arguments as a list of columns that pandas should convert to dates. This might replace your second to last line.
• index_col : This allows you to explicitly set an index, which should help you with not having to convert .to_frame()

If these don't get you all the way there, I have two ideas:

1. You can put all this logic inside of it's own function called something like clean_df and call that on newly loaded data.
2. You can save your cleaned data in a format other than a .csv . One (of many) option is that DataFrames can be saved to something called a pickle , which is one way python objects can be saved to memory . Loading DataFrames from a pickle brings them back pretty much exactly how you saved them, no need to do all the cleaning.

Also small note, I don't think you need to import pandas twice.