Formatting Data

All Bingo equations expect data to be formatted based on the number of variables and datapoints in the dataset.

Input

Bingo expects that input data is formatted with each variable as a column and each datapoint as a row.

Layout of inputs:

\(i\)	\(X_0\)	\(X_1\)	\(\ldots\)	\(X_n\)
0	0.1	1.2	\(\ldots\)	1.2
1	0.1	2.3	\(\ldots\)	3.5
2	0.1	1.2	\(\ldots\)	6.0
\(\vdots\)	\(\vdots\)	\(\vdots\)	\(\vdots\)	\(\vdots\)

Note

Bingo starts counting at 0, so \(X_0\) is the first variable, \(X_1\) is the second, and so on.

So, if we had 2 variables and 10 samples, we would have an array with 10 rows and 2 columns:

import numpy as np
X_0 = np.linspace(1, 10, num=10).reshape((-1, 1))
X_1 = np.linspace(-10, 1, num=10).reshape((-1, 1))
X = np.hstack((X_0, X_1))

Output

Bingo expects output data to be formatted as a of the same number of samples as the input.

Layout of output:

\(i\)	\(0\)	\(1\)	\(\ldots\)	\(n\)
\(y_i\)	0.0	-1.1	\(\ldots\)	5.0

Using the previous setup, let’s create output data by using the equation \(5.0 * X_0 + X_1\):

y = 5.0 * X_0 + X_1