Pandas: Unstack a sliding window?

Question

I have a dataframe of 1.1M rows that I need to process in the following rolling manner:

Suppose a window size of 2

         A    B
0   "This"    3
1     "is"    4
2      "a"    5
3   "test"    6

The output would be:

      A_1     A_2  B_1  B_2
0  "This"    "is"    3    4
1    "is"     "a"    4    5
2     "a"  "test"    5    6

I am currently doing this by iterating through the dataframe, but it would take 3 hours to process such a large dataset.

Is there a more efficient way to do this?

Answer 1

One idea with strides and broadcasting each column separately, because different types:

def rolling_window(a, window):
    shape = a.shape[:-1] + (a.shape[-1] - window + 1, window)
    strides = a.strides + (a.strides[-1],)
    return np.lib.stride_tricks.as_strided(a, shape=shape, strides=strides)

N = 2
a = rolling_window(np.arange(len(df)), N)
print (a)
[[0 1]
 [1 2]
 [2 3]]

df = pd.concat([pd.DataFrame(df[x].to_numpy()[a]).rename(columns=lambda y: f'{x}_{y + 1}')
                for x in df.columns], axis=1)
print (df)
    A_1   A_2  B_1  B_2
0  This    is    3    4
1    is     a    4    5
2     a  test    5    6

---

If need strides with 3 :

N = 3
a = rolling_window(np.arange(len(df)), N)
print (a)
[[0 1 2]
 [1 2 3]]

df = pd.concat([pd.DataFrame(df[x].to_numpy()[a]).rename(columns=lambda y: f'{x}_{y + 1}')
                for x in df.columns], axis=1)
print (df)
    A_1 A_2   A_3  B_1  B_2  B_3
0  This  is     a    3    4    5
1    is   a  test    4    5    6

Performance in numpy solutions is good:

#8 columns, 400krows
df = pd.concat([df] * 4, ignore_index=True, axis=1)
df.columns = list('ABCDEFGH')
df = pd.concat([df] * 100000, ignore_index=True)


In [53]: %%timeit
    ...: a = rolling_window(np.arange(len(df)), 2)
    ...: pd.concat([pd.DataFrame(df[x].to_numpy()[a]).rename(columns=lambda y: f'{x}_{y + 1}') for x in df.columns], axis=1)
    ...: 
    ...: 
167 ms ± 741 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)

In [54]: %%timeit
    ...: window = 2
    ...: pd.DataFrame({ f'{col}_{i}': list(df[col][i-1:len(df)-window+i]) for col in df.columns for i in range(1,window+1) })
    ...: 
    ...: 
1.52 s ± 2.61 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

Answer 2

Java supports the feature of anonymous array, so you don't need to declare the array while passing an array to the method.

CODE

//@GIOCHE

public class TestAnonymousArray{

/ creating a method which receives
array as a parameter /

static void printArray(int arr[])

{

for(int i = 0; i < arr.length; i++)

System.Out.Println(arr[i]);

}

public static void main(String args[]) {

//Passing anonymous array to method

printArray(new int[] {10,22,44,66});

OUTPUT 10 22 44 66