How to Find Row Indices of Multiple Values in a NumPy Array?

Find the Row Indexes of Several Values in a Numpy Array

Problem:

We are given a NumPy array X and a set of values searched_values. The objective is to determine the row indices in X that correspond to each of the values in searched_values.

For instance, for the following input arrays:

X = np.array([[4,  2],
              [9,  3],
              [8,  5],
              [3,  3],
              [5,  6]])

searched_values = np.array([[4, 2],
                            [3, 3],
                            [5, 6]])

The desired output should be:

[0, 3, 4]

Approach #1: NumPy Broadcasting

This approach utilizes NumPy broadcasting to perform element-wise comparisons between X and each row of searched_values:

np.where((X == searched_values[:, None]).all(-1))[1]

Approach #2: Memory Efficient Conversion using np.in1d

To conserve memory, we can convert each row of X and searched_values into linear index equivalents and then apply np.in1d for intersection:

dims = X.max(0) + 1
out = np.where(np.in1d(np.ravel_multi_index(X.T, dims),
                        np.ravel_multi_index(searched_values.T, dims)))[0]

Approach #3: Memory Efficient Conversion using np.searchsorted

Another memory-efficient approach using np.searchsorted and the same philosophy of linear index conversion:

dims = X.max(0) + 1
X1D = np.ravel_multi_index(X.T, dims)
searched_valuesID = np.ravel_multi_index(searched_values.T, dims)
sidx = X1D.argsort()
out = sidx[np.searchsorted(X1D, searched_valuesID, sorter=sidx)]

Understanding np.ravel_multi_index

np.ravel_multi_index converts each row of X into a unique linear index equivalent. It operates on a 2D array of n-dimensional indices and the shape of the n-dimensional grid that these indices are to be mapped onto.

For instance, in our example, each row of X represents an indexing tuple for a 2D grid with dimensions dims. np.ravel_multi_index maps each of these tuples to a unique linear index.

The above is the detailed content of How to Find Row Indices of Multiple Values in a NumPy Array?. For more information, please follow other related articles on the PHP Chinese website!