01. Sliding window

Out:

Data:
      feature_0  feature_1  feature_2  feature_3  feature_4  feature_5  feature_6  feature_7  feature_8  feature_9  patient  day
0      0.782839   0.579171   0.875666   0.896433   0.522666   0.424833   0.019499   0.843256   0.508497   0.023651        0  -15
1      0.938406   0.686556   0.716277   0.474455   0.839579   0.552715   0.597967   0.308887   0.613897   0.567229        0  -14
2      0.917493   0.842386   0.728984   0.522605   0.850795   0.407061   0.984394   0.082715   0.594869   0.059661        0  -13
3      0.722237   0.947399   0.549326   0.725588   0.976665   0.596684   0.882883   0.594011   0.760227   0.505716        0  -12
4      0.680577   0.052478   0.142805   0.977845   0.755283   0.091267   0.700849   0.692678   0.687108   0.046146        0  -11
...         ...        ...        ...        ...        ...        ...        ...        ...        ...        ...      ...  ...
1495   0.668256   0.162212   0.051996   0.768991   0.126166   0.594980   0.606882   0.973683   0.444446   0.438552       99   -5
1496   0.074663   0.669059   0.815666   0.425611   0.900212   0.069323   0.597965   0.300225   0.087881   0.892852       99   -4
1497   0.572242   0.856386   0.745518   0.044273   0.610881   0.320131   0.804878   0.652734   0.014684   0.986622       99   -3
1498   0.266470   0.857181   0.888199   0.421400   0.169874   0.498681   0.395948   0.240385   0.194971   0.757166       99   -2
1499   0.131035   0.197843   0.184269   0.605561   0.850481   0.687066   0.508127   0.683124   0.304230   0.779854       99   -1

[1500 rows x 12 columns]
C:\Users\kelda\Desktop\repositories\github\python-spare-code\main\examples\pandas\plot_format01.py:50: FutureWarning:

DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.


Result:
              feature_0  feature_1  feature_2  feature_3  feature_4  feature_5  feature_6  feature_7  feature_8  feature_9  patient  day  window
patient
0       0      0.782839   0.579171   0.875666   0.896433   0.522666   0.424833   0.019499   0.843256   0.508497   0.023651        0  -15       0
        1      0.938406   0.686556   0.716277   0.474455   0.839579   0.552715   0.597967   0.308887   0.613897   0.567229        0  -14       0
        2      0.917493   0.842386   0.728984   0.522605   0.850795   0.407061   0.984394   0.082715   0.594869   0.059661        0  -13       0
        1      0.938406   0.686556   0.716277   0.474455   0.839579   0.552715   0.597967   0.308887   0.613897   0.567229        0  -14       1
        2      0.917493   0.842386   0.728984   0.522605   0.850795   0.407061   0.984394   0.082715   0.594869   0.059661        0  -13       1
...                 ...        ...        ...        ...        ...        ...        ...        ...        ...        ...      ...  ...     ...
99      1497   0.572242   0.856386   0.745518   0.044273   0.610881   0.320131   0.804878   0.652734   0.014684   0.986622       99   -3      13
        1498   0.266470   0.857181   0.888199   0.421400   0.169874   0.498681   0.395948   0.240385   0.194971   0.757166       99   -2      13
        1497   0.572242   0.856386   0.745518   0.044273   0.610881   0.320131   0.804878   0.652734   0.014684   0.986622       99   -3      14
        1498   0.266470   0.857181   0.888199   0.421400   0.169874   0.498681   0.395948   0.240385   0.194971   0.757166       99   -2      14
        1499   0.131035   0.197843   0.184269   0.605561   0.850481   0.687066   0.508127   0.683124   0.304230   0.779854       99   -1      14

[3900 rows x 13 columns]

# Interesting code.
# np.lib.stride_tricks.sliding_window_view(df.index, 3)

# Libraries
import numpy as np
import pandas as pd

# Configuration
ROWS, COLS = 1500, 10
PATIENTS = 100

# Create random values
features = np.random.random_sample((ROWS, COLS))
patients = np.random.randint(PATIENTS, size=(ROWS,1))

# Create DataFrame
df = pd.DataFrame(data=features)
df = df.add_prefix('feature_')
df['patient'] = patients
df['day'] = -(df.groupby('patient').cumcount()+1)
df = df.sort_values(by=['patient', 'day'],
    ascending=[True, True]).reset_index(drop=True)

# Show
print("\nData:")
print(df)

# ----------------------------------
# Method I: Own method
# ----------------------------------
def sliding_window_iter(series, size, include_id=True):
    """series is a column of a DataFrame.

    .. note: The DataFrame should be pre-ordered to ensure
             that IDs remain consistent.
    """
    for i, start_row in enumerate(range(len(series) - size + 1)):
        s = series[start_row:start_row + size]
        if include_id:
            s['window'] = i
        yield s


# Group by patient and compute sliding window
result = df.groupby('patient').apply(lambda x:
    pd.concat(sliding_window_iter(x, 3)))

# Show
print("\nResult:")
print(result)

# ----------------------------------
# Method II: Using rolling
# ----------------------------------
#a = df.groupby('patient').rolling(window=3)
#b = [win for win in a if win.shape[0] == 3]
#c = pd.concat(b)
#print(c)