Plotting the count of values per week from a Dataframe

Question

Say I have a pandas dataframe of the following format illustrating the total purchase value ($) of a person in a store at some time stamp for everyday in the year of 2015:

       person_num               date       value
0             1  2015-01-01 00:23:32       298.00
1             2  2015-01-01 00:31:59       348.00
2             3  2015-01-01 02:30:21       533.00
3             4  2015-01-01 03:14:10       022.00
4             5  2015-01-01 05:51:43       448.00
5             6  2015-01-01 07:10:08       158.00
...         ...                  ...          ...
123728   123729  2015-12-31 23:13:34       109.00

etc. and this goes up to 2015-12-31... How would I be able to plot the total value per week. So the plot would have the week number on the x-axis and total value (sum of all the purchase amounts per week ) on y-axis.

Edit: This pertains only to the value column not to the person num

Answer 1

import pandas as pd

# Construct input dataframe
df = pd.DataFrame({
    "person_num": [
        1,
        2,
        3,
        4,
        5,
        6,
    ],
    "date": [
        "2015-01-01 00:23:32",
        "2015-01-02 00:31:59",
        "2015-01-05 02:30:21",
        "2015-01-06 03:14:10",
        "2015-01-15 05:51:43",
        "2015-01-16 07:10:08",
    ],
    "value": [
        298.00,
        348.00,
        533.00,
        022.00,
        448.00,
        158.00,
    ],
})

# Construct dataframe with week number column and values column
df2 = df.loc[:, ["value"]]
df2["week_no"] = pd.to_datetime(df["date"]).dt.week

# Sum up weekly values
df3 = (
    df2
    .groupby("week_no")
    .sum()
    .reset_index()
)

# Plot results
df3.plot.scatter(x="week_no", y="value")