如何按纬度在 90 到 -90 范围内拆分 dataframe

Question

I have dataframe that I need to split based on latitude.

The example of dataframe columns:

df.glat = [-20, -30, -40, -50, -60, -70, -80, -90, -80, -70 - 60, -50, -40, -30, -20, -10, 10, 20, 30]

I was trying to create a column status based on behavior of latitude values:

df. status = ['decline', 'decline', 'decline', 'decline', 'decline', 'decline', 'decline', 'decline', 'increase', 'increase', 'increase', 'increase', 'increase', 'increase', 'increase', 'increase', 'increase', 'increase', 'increase']

I need to group it into groups from 90 to -90. In original my dataframe change in this ranges several time.

How can split dataframes into such groups:

1 : [-20, -30, -40, -50, -60, -70, -80, -90]
2: [ -80, -70 - 60, -50, -40, -30, -20, -10, 10, 20, 30]

I have made a graphic of my ordinal data how the latitude changes: So I need to split each range from [-90 to 90] as separate dataframe. Like here from the graphic for example should be 16 splitted dataframes.

I will be grateful for any suggestion!

Answer 1

df = pd.DataFrame({'glat': [-20, -30, -40, 10, 20, 30, -40, -50, -60]})

df['is_increasing'] = df['glat'].diff().bfill().gt(0)
df['group'] = df['is_increasing'].diff().bfill().cumsum()

df.groupby('group')['glat'].apply(list)

Output:

group
0    [-20, -30, -40]
1       [10, 20, 30]
2    [-40, -50, -60]
Name: glat, dtype: object

PS df will get a group column with group ID, so you can easily split it into separate dataframes later (for example, with dataframes = [df for _, df in df.groupby('group')] ):

   glat  is_increasing  group
0   -20          False      0
1   -30          False      0
2   -40          False      0
3    10           True      1
4    20           True      1
5    30           True      1
6   -40          False      2
7   -50          False      2
8   -60          False      2