Python Pandas - 时间/日期、平均数据和绘制图形

Question

I am having trouble with preparing data so that it is suitable to be plotted to a graph. This is my dataframe:

                          Date  Lane Lane Name  Direction DirectionName  \
0      2018-02-02 00:00:03.000     6     SB_NS          2         South   
1      2018-02-02 00:00:22.010     5    SB_MID          2         South   
2      2018-02-02 00:00:22.020     4     SB_OS          2         South   
3      2018-02-02 00:00:36.040     6     SB_NS          2         South   
4      2018-02-02 00:00:49.070     6     SB_NS          2         South   
...                        ...   ...       ...        ...           ...   
503763 2018-02-27 23:59:00.090     2    NB_MID          1         North   
503764 2018-02-27 23:59:29.090     6     SB_NS          2         South   
503765 2018-02-27 23:59:32.050     4     SB_OS          2         South   
503766 2018-02-27 23:59:33.070     6     SB_NS          2         South   
503767 2018-02-27 23:59:58.050     1     NB_NS          1         North   

        Speed (mph)  Headway (s)  Gap (s)  Flags Flag Text  
0            38.525          NaN      NaN      5    Friday  
1            32.310          NaN      NaN      5    Friday  
2            44.739          NaN      NaN      5    Friday  
3            33.554          NaN      NaN      5    Friday  
4            39.768       12.300   11.847      5    Friday  
...             ...          ...      ...    ...       ...  
503763       32.932        4.415    3.833      2   Tuesday  
503764       29.825       65.500   64.700      2   Tuesday  
503765       29.205      236.000  235.848      2   Tuesday  
503766       37.283        3.330    3.462      2   Tuesday  
503767       36.661       76.000   75.669      2   Tuesday  

[503768 rows x 10 columns]

It is traffic data. Each row is a single observation of traffic at a point in time. Flags is simply the day of the week. The data has been collected on every Tuesday and Friday of the month. So the dataframe contains 8 different dates, 4 Tuesdays, 4 Fridays

I want to plot two graphs. One graph will show only South data, the other will show only north data. Both graphs should show average traffic volume for each hour of the day on a selected date of my choice ( 2018-02-02 for example).

So to clarify, here is what the output should be:

• Two bar plots, one for the North direction and one for South, for 2018-02-02

• Each bar plot should show the average traffic volume for each hour of the day.

I am just a bit confused about how to only collect data for a single date and how to collect the average traffic flow for that date per hour.

So far, I have grouped by date/hour and counted the total... As shown below.

Date    DirectionName   count
0   2018-02-02 00:00:00 North   212
1   2018-02-02 00:00:00 South   250
2   2018-02-02 01:00:00 North   130
3   2018-02-02 01:00:00 South   137
4   2018-02-02 02:00:00 North   76
... ... ... ...
379 2018-02-27 21:00:00 South   801
380 2018-02-27 22:00:00 North   425
381 2018-02-27 22:00:00 South   511
382 2018-02-27 23:00:00 North   233
383 2018-02-27 23:00:00 South   301

The problem is, the count is obviously not an average per hour. This method also uses every single date, when I only want to use a specific date, such as 2018-02-02.

• How do I change my current method to show an average per hour instead of total per hour?
• How do I change my current method to show ONLY a specific date?
• Is my current method unsuitable / is there a better method?

This is the code for my current method;

df.Date=pd.to_datetime(df.Date)
df.groupby([pd.Grouper(key='Date',freq='H'),df.DirectionName]).size().reset_index(name='count')

Some advice / Clarification would be greatly appreciated :)

Answer 1

Filter the data using .loc to only a single day, then you are correctly counting the number of rows per hour, after which simply group by day and get the average value.

df.Date=pd.to_datetime(df.Date)
df = df[df["Date"] == "2018-02-02"]
hourly = df.groupby([pd.Grouper(key='Date',freq='H'),df.DirectionName]).size().reset_index(name='count')
daily = df.groupby([df.Date, df.DirectionName]).mean()

This gives you the average amount of traffic per hour on that day. It returns a single number - I am not sure if that is what you want. If not, do you want one of the other fields averaged? eg average speed per hour? for every single hour of the day?