Sparse SVD (svds) in SciPy - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the sparse SVD function from scipy.

SciPy

from scipy.sparse.linalg import [1]

Drag options to blanks, or click blank then click option'

Asparse_svd_solver

Bsvd

Csparse_svd

Dsvds

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'svd' which is for dense matrices.

Trying to import a non-existent function like 'sparse_svd'.

The function svds is the correct sparse SVD function in scipy.sparse.linalg.

2fill in blank

medium

Complete the code to create a sparse matrix using scipy.

SciPy

from scipy.sparse import [1]
data = [1]((3, 3))

Drag options to blanks, or click blank then click option'

Acoo_matrix

Bcsr_matrix

Ccsc_matrix

Ddense_matrix

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'dense_matrix' which does not exist in scipy.sparse.

Choosing 'csc_matrix' which is column-based, not row-based.

csr_matrix creates a compressed sparse row matrix, commonly used for sparse data.

3fill in blank

hard

Fix the error in the code to compute the top 2 singular values and vectors of a sparse matrix.

SciPy

u, s, vt = svds(data, k=[1])

Drag options to blanks, or click blank then click option'

A2

B3

C1

D0

Attempts:

3 left

💡 Hint

Common Mistakes

Using k=3 when the matrix dimension is smaller.

Using k=0 which is invalid.

The parameter k specifies the number of singular values and vectors to compute. For top 2, use k=2.

4fill in blank

hard

Fill both blanks to create a sparse matrix and compute its sparse SVD with 1 singular value.

SciPy

from scipy.sparse import [1]
from scipy.sparse.linalg import svds
matrix = [1]([[1, 0], [0, 1]])
u, s, vt = svds(matrix, k=[2])

Drag options to blanks, or click blank then click option'

Acsr_matrix

Bcsc_matrix

C1

D2

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'csc_matrix' instead of 'csr_matrix'.

Setting k to 2 when only 1 singular value is needed.

Use csr_matrix to create the sparse matrix and k=1 to compute one singular value.

5fill in blank

hard

Fill all three blanks to compute sparse SVD and reconstruct the original matrix.

SciPy

u, s, vt = svds(matrix, k=[1])
S = np.diag(s)
reconstructed = u [2] S [3] vt

Drag options to blanks, or click blank then click option'

A2

B@

C*

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using '*' which does element-wise multiplication, not matrix multiplication.

Setting k to 2 when only 1 singular value is needed.

Use k=1 for one singular value. Matrix multiplication is done with the @ operator in Python for arrays.