Pandas DataFrame-如何制作堆积面积图堆栈（matplotlib）

Question

I am trying to convert data in a pandas DataFrame in to a stacked area graph but can not seem to get it to stack.

The data is in the format

index | datetime (yyyy/mm/dd) | name | weight_change

With 6 different people each measured daily.

I want the stacked graph to show the weight_change (y) over the datetime (x) but with weight_change for each of the 6 people stacked on top of each other

The closest I have been able to get to it is with:

df = df.groupby['datetime', 'name'], as_index=False).agg({'weight_change': 'sum'})
agg = df.groupby('datetime').sum()
agg.plot.area()

This produces the area graph for the aggregate of the weight_change values (sum of each persons weight_change for each day) but I can't figure out how to split this up for each person like the different values here:

I have tried various things with no luck. Any ideas?

Answer 1

A simplified version of your data:

df = pd.DataFrame(dict(days=range(4)*2,
                       change=np.random.rand(8)*2.,
                       name=['John',]*4 + ['Jane',]*4))

df :

    change  days    name
0   0.238336    0   John
1   0.293901    1   John
2   0.818119    2   John
3   1.567114    3   John
4   1.295725    0   Jane
5   0.592008    1   Jane
6   0.674388    2   Jane
7   1.763043    3   Jane

Now we can simply use pyplot 's stackplot :

import matplotlib.pyplot as plt

days = df.days[df.name == 'John']
plt.stackplot(days, df.change[df.name == 'John'],
              df.change[df.name == 'Jane'])

This produces the following plot: