Matrix Math

Question 1

A single value is known as

Answer 1

scalar

Question 2

scalar

Answer 2

a value with 0 dimensions

Question 3

Lists of values are known as

Answer 3

vectors

Question 4

Types of vectors

Answer 4

row and column

Question 5

Dimensions of vectors

Answer 5

just one: length

Question 6

What is a matrix

Answer 6

a 2 dimensional grid of values

Question 7

What is a tensor?

Answer 7

Any n-dimensional collection of values

Question 8

Locations in matrices are known as

Answer 8

indices

Question 9

Indices nomenclature

Answer 9

Like a 1 layer nested array

Question 10

Numpy is

Answer 10

a C library in python.

Does lots of math operations in Python and is designed to work with matrices.

Question 11

Normal convention for naming numpy

Answer 11

import numpy as np

Question 12

Most common way to with number in NumPy is through

Answer 12

ndarray objects

Question 13

ndarray objects are

Answer 13

similar to Python lists, but can have any number of dimensions
Does fast math operations

Question 14

To declare an ndarray

Answer 14

x = nd.array(5)

Question 15

To get shape of ndarray

Answer 15

nd.shape

Question 16

To reshape an nd array like one that is (4,)

Answer 16

(4,).reshape(1,4)

Question 17

Why do some people use

x = v[:, None]

Answer 17

Adds extra dimension

Question 18

Elementwise operations

Answer 18

Like iterating through and running an operation

Question 19

Requirements for adding two matrices

Answer 19

Have to be the same shape

Question 20

When describing the shape of a matrix how does one describe it?

Answer 20

rows x columns

Question 21

You can only safely run a transpose to multiply if

Answer 21

The data is arranged as rows

Question 22

To get the min, max, mean of a matrix

Answer 22

np. min(array)
np. max(array)
mp. mean(array)

Question 23

How to calculate error in a logistic regression?

Answer 23

It the number of errors

Question 24

What method does one use to minimize the error?

Answer 24

Gradient descent

Question 25

Basic parts of a neural network

Answer 25

Input data, processing, output

Question 26

Individual nodes are called

Answer 26

perceptrons

Question 27

What are weights?

Answer 27

A higher weight means the neural network considers that input more important than other inputs, and lower weight means that the data is considered less important.

Question 28

W vs w

Answer 28

W when it represents a matrix of weights or a w when it represents an individual weight

Question 29

How is an output signal determined?

Answer 29

feeding the linear combination into an activation function

Question 30

What are two ways to go from an AND perceptron to an OR perceptron?

Answer 30

Increase the weights

Decrease the magnitude of the bias

Question 31

AND perceptron

Answer 31

Both must be true to accept

Question 32

OR perceptron

Answer 32

One must be true

Question 33

NOT perceptron

Answer 33

A specific one must be true

Question 34

XOR perceptron

Answer 34

outputs 0 if the inputs are the same and 1 if the inputs are different

Question 35

Gradient is

Answer 35

term for rate of change or slope

Question 36

To calculate rate of change

Answer 36

derivative of a function f(x) gives you another function f​’(x) that returns the slope of f(x) at point x

Question 37

Local minima

Answer 37

where the error is low, but not the lowest

Question 38

SSE is

Answer 38

measure of networks performance. Low means good predictions.

Question 39

np.dot is the same as

Answer 39

Multiplying two matrices and then getting the sum

Question 40

sigmoid(x)

Answer 40

1/(1+np.exp(-x))

Question 41

sigmoid_prime(x)

Answer 41

sigmoid(x) * (1 - sigmoid(x))