如何计算和使用 python 从调查中收到的 plot 数据

Question

What I need to do is basically calculate the responses received over a period of time.

IE

07/07/2019 |  6
08/07/2019 |  7

And plot the above to a graph.

But the current data is in the below format:

07/07/2019 17:33:07
07/07/2019 12:00:03
08/07/2019 21:10:05
08/07/2019 20:06:09

So far,

import pandas as pd
df = pd.read_csv('survey_results_public.csv')
df.head()
df['Timestamp'].value_counts().plot(kind="bar")
plt.show()

But the above doesn't look good.

Answer 1

You are counting all values in the timestamp column so you will have 1 response per timestamp.

You should parse the timestamp column, check the unique dates and then count the number of timestamps that belong to each date.

Only then should you plot the data.

So do something like this:

import pandas as pd
import datetime

def parse_timestamps(timestamp):
    datetime.datetime.strptime(timestamp, '%d/%m/%Y %H:%M:%S')

df = pandas.read_csv('survey_results_public.csv')

df["Date"]=df["Timestamp"].map(lambda t: parse_timestamps(t).date())

df["Date"].value_counts().plot(kind="bar")