Happy to suggest a (possibly obvious) implementation of this, which could be made in pure Python or C/Cython if you’ve got time and space for new dependencies, and need it to be faster. A sparse matrix in N dimensions can assume most elements are empty, so we use a dictionary keyed on tuples: class NDSparseMatrix: … Read more
You can pass a numpy array or matrix as an argument when initializing a sparse matrix. For a CSR matrix, for example, you can do the following. >>> import numpy as np >>> from scipy import sparse >>> A = np.array([[1,2,0],[0,0,3],[1,0,4]]) >>> B = np.matrix([[1,2,0],[0,0,3],[1,0,4]]) >>> A array([[1, 2, 0], [0, 0, 3], [1, 0, … Read more
You can use tf.where and tf.gather_nd to do that: import numpy as np import tensorflow as tf # Make a tensor from a constant a = np.reshape(np.arange(24), (3, 4, 2)) a_t = tf.constant(a) # Find indices where the tensor is not zero idx = tf.where(tf.not_equal(a_t, 0)) # Make the sparse tensor # Use tf.shape(a_t, out_type=tf.int64) … Read more
I don’t see what the advantages of csr format are in this case. Sure, all the nonzero values are collected in one .data array, with the corresponding column indexes in .indices. But they are in blocks of varying length. And that means they can’t be processed in parallel or with numpy array strides. One solution … Read more
Edit: bbtrb’s method (using coo_matrix) is much faster than my original suggestion, using nonzero. Sven Marnach’s suggestion to use itertools.izip also improves the speed. Current fastest is using_tocoo_izip: import scipy.sparse import random import itertools def using_nonzero(x): rows,cols = x.nonzero() for row,col in zip(rows,cols): ((row,col), x[row,col]) def using_coo(x): cx = scipy.sparse.coo_matrix(x) for i,j,v in zip(cx.row, cx.col, … Read more
You can use the scipy.sparse.hstack to concatenate sparse matrices with the same number of rows (horizontal concatenation): from scipy.sparse import hstack hstack((X, X2)) Similarly, you can use scipy.sparse.vstack to concatenate sparse matrices with the same number of columns (vertical concatenation). Using numpy.hstack or numpy.vstack will create an array with two sparse matrix objects.