Numpy

Question 1

reshape (len(y)L,) vector into (len(y)L, 1L)

Answer 1

y = y.reshape((len(y),1))

Question 2

column wize means and standard deviations of X np array

Answer 2

means, stds = X.mean(axis=0), X.std(axis=0)

Question 3

adding column of 1s to np array X (2ways)

Answer 3

X = np.hstack((np.ones((len(X[:,0]),1)),X))
X = np.c_[np.ones((len(X[:,0]),1)),X]

Question 4

median of y vector

Answer 4

np.median(y)

Question 5

matrix multiplication for transposed X and X

Answer 5

np.dot(X.T,X)

Question 6

solve X * v = y for v

Answer 6

v = np.linalg.solve(X, y)

Question 7

creating pseudorandom number

Answer 7

np.random.seed(seed)

random_ind = np.random.randint(X.shape[0])

Question 8

calculate euclidean distance between w1 and w0

Answer 8

weight_dist = np.linalg.norm(w1 - w0)

Question 9

create [4,1] np array with zeros

Answer 9

w_init = np.zeros((4, 1))

Question 10

create np array from list

and list from np.array

Answer 10

some_array = np.array(some_list)

Question 11

column wize sum of elements in np_array

Answer 11

np_array.sum(axis=0)

Question 12

number of rows in np_array

Answer 12

np_array.shape[0]

Question 13

flatten array a

equivalent to a.reshape(-1)

Answer 13

np.ravel(a,order=’C’)
1)possible orders: {‘C’,’F’, ‘A’, ‘K’}
‘C’ means to index the elements in row-major
‘F’ means to index the elements in column-major
‘A’ means to read the elements in Fortran-like index order if a is Fortran contiguous in memory, C-like order otherwise
‘K’ means to read the elements in the order they occur in memory, except for reversing the data when strides are negative

Question 14

How to count the occurrence of certain item in an ndarray in Python

Answer 14

unique, counts = numpy.unique(a, return_counts=True)

dict(zip(unique, counts))

Question 15

Sort np.array a in descending order

Answer 15

-np.sort(-a)

Question 16

correlation coefficient between numpy list1 and list2

Answer 16

numpy.corrcoef(list1, list2)[0, 1]

Question 17

index of maximum element in numpy array

Answer 17

a.argmax(axis=0)

Question 18

Convert predictions into categories

Answer 18

import numpy as np

preds = [1.22, 3.32, 6.54, 3.22, 6.78, 8.1, 2.01]
splits = [0, 1.5, 2.5, 3, 4.2, 5.8, 6.5, 7]

response = np.digitize(preds, splits)

Question 19

add vectors of different sizes

Answer 19

placebo_zeros = zeros((11034))
placebo_ones = ones((189))
placebo_ones.resize(placebo_zeros.shape)

placebo_data = placebo_zeros + placebo_ones