Numpy and Scipy

Numerical Computing in Python

1
What is Numpy?
• Numpy, Scipy, and Matplotlib provide MATLAB-like
functionality in python.
• Numpy Features:
 Typed multidimentional arrays (matrices)
 Fast numerical computations (matrix math)
 High-level math functions

2
Why do we need NumPy
Let’s see for ourselves!

3
Why do we need NumPy
• Python does numerical computations slowly.
• 1000 x 1000 matrix multiply
 Python triple loop takes > 10 min.
 Numpy takes ~0.03 seconds

4
Logistics: Versioning
• In this class, your code will be tested with:
 Python 2.7.6
 Numpy version: 1.8.2
 Scipy version: 0.13.3
 OpenCV version: 2.4.8

• Two easy options:

 Class virtual machine (always test on the VM)
 Anaconda 2 (some assembly required)

5
NumPy Overview
1. Arrays
2. Shaping and transposition
3. Mathematical Operations
4. Indexing and slicing
5. Broadcasting

6
Arrays
Structured lists of numbers.
• Vectors

• Matrices

• Images

• Tensors

• ConvNets

7
Arrays
Structured lists of numbers. • 
• Vectors

• Matrices

• Images

• Tensors

• ConvNets

8
Arrays
Structured lists of numbers.
• Vectors

• Matrices

• Images

• Tensors

• ConvNets

9
Arrays
Structured lists of numbers.
• Vectors

• Matrices

• Images

• Tensors

• ConvNets

10
Arrays
Structured lists of numbers.
• Vectors

• Matrices

• Images

• Tensors

• ConvNets

11
Arrays, Basic Properties
import numpy as np
a = np.array([[1,2,3],[4,5,6]],dtype=np.float32)
print a.ndim, a.shape, a.dtype

1. Arrays can have any number of dimensions, including zero (a scalar).

2. Arrays are typed: np.uint8, np.int64, np.float32, np.float64
3. Arrays are dense. Each element of the array exists and has the same type.

12
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

13
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

14
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

15
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

16
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

17
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

18
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

19
Arrays, creation
• np.ones, np.zeros
• np.arange

• np.concatenate

• np.astype

• np.zeros_like, np.ones_like
• np.random.random

20
Arrays, danger zone
• Must be dense, no holes.
• Must be one type
• Cannot combine arrays of different shape

21
Shaping
a = np.array([1,2,3,4,5,6])
a = a.reshape(3,2)
a = a.reshape(2,-1)
a = a.ravel()
1. Total number of elements cannot change.
2. Use -1 to infer axis shape
3. Row-major by default (MATLAB is column-major)

22
Return values
• Numpy functions return either views or copies.
• Views share data with the original array, like references
in Java/C++. Altering entries of a view, changes the
same entries in the original.
• Thenumpy documentation says which functions return
views or copies
• Np.copy, np.view make explicit copies and views.

23
Transposition
a = np.arange(10).reshape(5,2)
a = a.T
a = a.transpose((1,0))

np.transpose permutes axes.

a.T transposes the first two axes.

24
Saving and loading arrays
np.savez(‘data.npz’, a=a)
data = np.load(‘data.npz’)
a = data[‘a’]

1. NPZ files can hold multiple arrays

2. np.savez_compressed similar.

25
Image arrays
Images are 3D arrays: width, height, and channels
Common image formats:
height x width x RGB (band-interleaved)
height x width (band-sequential)

Gotchas:
Channels may also be BGR (OpenCV does this)
May be [width x height], not [height x width]
26
Saving and Loading Images
SciPy: skimage.io.imread,skimage.io.imsave
height x width x RGB
PIL / Pillow: PIL.Image.open, Image.save
width x height x RGB
OpenCV: cv2.imread, cv2.imwrite
height x width x BGR

27
Recap
We just saw how to create arrays, reshape them, and
permute axes

Questions so far?

28
Recap
We just saw how to create arrays, reshape them, and
permute axes

Questions so far?

Now: let’s do some math

29
Mathematical operators
• Arithmetic operations are element-wise
• Logical operator return a bool array
• In place operations modify the array

30
Mathematical operators
• Arithmetic operations are element-wise
• Logical operator return a bool array
• In place operations modify the array

31
Mathematical operators
• Arithmetic operations are element-wise
• Logical operator return a bool array
• In place operations modify the array

32
Mathematical operators
• Arithmetic operations are element-wise
• Logical operator return a bool array
• In place operations modify the array

33
Math, upcasting
Just as in Python and Java, the result of a math operator is cast
to the more general or precise datatype.
uint64 + uint16 => uint64
float32 / int32 => float32

Warning: upcasting does not prevent overflow/underflow. You

must manually cast first.
Use case: images often stored as uint8. You should convert to
float32 or float64 before doing math.
34
Math, universal functions
Also called ufuncs
Element-wise
Examples:
 np.exp
 np.sqrt
 np.sin
 np.cos
 np.isnan

35
Math, universal functions
Also called ufuncs
Element-wise
Examples:
 np.exp
 np.sqrt
 np.sin
 np.cos
 np.isnan

36
Math, universal functions
Also called ufuncs
Element-wise
Examples:
 np.exp
 np.sqrt
 np.sin
 np.cos
 np.isnan

37
Indexing
x[0,0] # top-left element
x[0,-1] # first row, last column
x[0,:] # first row (many entries)
x[:,0] # first column (many entries)
Notes:
 Zero-indexing
 Multi-dimensional indices are comma-separated (i.e., a tuple)

38
Indexing, slices and arrays
I[1:-1,1:-1] # select all but one-pixel border
I = I[:,:,::-1] # swap channel order
I[I<10] = 0 # set dark pixels to black
I[[1,3], :] # select 2nd and 4th row

1. Slices are views. Writing to a slice overwrites the original array.

2. Can also index by a list or boolean array.

39
Python Slicing
Syntax: start:stop:step
a = list(range(10))
a[:3] # indices 0, 1, 2
a[-3:] # indices 7, 8, 9
a[3:8:2] # indices 3, 5, 7
a[4:1:-1] # indices 4, 3, 2 (this one is tricky)

40
Axes
a.sum() # sum all entries
a.sum(axis=0) # sum over rows
a.sum(axis=1) # sum over columns
a.sum(axis=1, keepdims=True)
1. Use the axis parameter to control which axis NumPy
operates on
2. Typically, the axis specified will disappear, keepdims
keeps all dimensions
41
Broadcasting

a = a + 1 # add one to every element

When operating on multiple arrays, broadcasting rules are used.

Each dimension must match, from right-to-left
1. Dimensions of size 1 will broadcast (as if the value was repeated).
2. Otherwise, the dimension must have the same shape.
3. Extra dimensions of size 1 are added to the left as needed.

42
Broadcasting example
Suppose we want to add a color value to an image
a.shape is 100, 200, 3
b.shape is 3
a + b will pad b with two extra dimensions so it has an
effective shape of 1 x 1 x 3.
So, the addition will broadcast over the first and second
dimensions.

43
Broadcasting failures
If a.shape is 100, 200, 3 but b.shape is 4 then a + b will
fail. The trailing dimensions must have the same shape (or
be 1)

44
Tips to avoid bugs
1. Know what your datatypes are.
2. Check whether you have a view or a copy.
3. Use matplotlib for sanity checks.
4. Use pdb to check each step of your computation.
5. Know np.dot vs np.mult.

45
Average images
Who is this?

46
Practice exercise (not graded)
Compute the average image of faces.
1. Download Labeled Faces in the Wild dataset (google:
LFW face dataset). Pick a face with at least 100 images.
2. Call numpy.zeros to create a 250 x 250 x 3 float64
tensor to hold the result
3. Read each image with skimage.io.imread, convert to
float and accumulate
4. Write the averaged result with skimage.io.imsave

47