Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Numerical tour in the Python eco-system 
Python, NumPy, scikit-learn 
Arnaud Joly 
October 2, 2014
2 / 37
How to install Python? 
Download and use the Anaconda python distribution 
https://store.continuum.io/cshop/anaconda/. It ...
Using the python interpreter 
Interactive mode 
1. Start a python shell 
$ ipython 
2. Write python code 
>>> print("Hello...
Basic types 
Integer >>> 5 
5 
>>> a = 5 
>>> a 
5 
Float >>> pi = 3.14 
complex >>> c = 1 - 1j 
boolean >>> b = 5 > 3 # 5...
Python is a dynamic program language 
Variable types are implicitly inferred during the assignment. 
Variables are not dec...
Numbers and their arithmetic operations (+,-,/,//,*,**,%) 
>>> 1 + 2 
4 
>>> 50 - 5 * 6 
20 
>>> 2 / 3 # with py3 0.66... ...
Playing with strings 
>>> s = ’Great day!’ 
>>> s 
’Great day!’ 
>>> s[0] # strings are sequences 
’G’ 
>>> """A very 
ver...
list, an ordered collection of objects 
Instantiation >>> l = [] # an empty list 
>>> l = [’spam’, ’egg’, [’another list’]...
dict, an unordered and associative data structure of 
key-value pairs 
Instantiation >>> d = {1: "a", "b": 2, 0: [4, 5, 6]...
dict, an unordered and associative data structure of 
key-value pairs 
Methods >>> len(d) 
3 
>>> d.keys() 
[0, 1, ’b’] 
>...
Control flow: if / elif / else 
>>> x = 3 
>>> if x == 0: 
... print("zero") 
... elif x == 1: 
... print("one") 
... else...
Control flow: for loop 
>>> l = [0, 1, 2, 3] 
>>> for a in l: # Iterate over a sequence 
... print(a ** 2) 
0 
1 
4 
Itera...
Control flow: while 
>>> a, b = 0, 1 
>>> while b < 50: # while True do ... 
... a, b = b, a + b 
... print(a) 
... 
1 
1 ...
Control flow: functions 
>>> def f(x, e=2): 
... return x ** e 
... 
>>> f(3) 
9 
>>> f(5, 3) 
125 
>>> f(5, e=3) 
125 
Fu...
Classes and object 
>>> class Counter: 
... def __init__(self, initial_value=0): 
... self.value = initial_value 
... def ...
Import a package 
>>> import math 
>>> math.log(3) 
1.0986122886681098 
>>> from math import log 
>>> log(4) 
1.3862943611...
Python reference and tutorial 
I Python Tutorial : http://docs.python.org/tutorial/ 
I Python Reference : https://docs.pyt...
19 / 37
NumPy 
NumPy is the fundamental package for scientific computing with 
Python. It contains among other things: 
I a powerf...
1-D numpy arrays 
Let’s import the package. 
>>> import numpy as np 
Let’s create a 1-dimensional array. 
>>> a = np.array...
2-D numpy arrays 
Let’s import the package. 
>>> import numpy as np 
Let’s create a 2-dimensional array. 
>>> b = np.array...
Array operations 
>>> a = np.ones(3) / 5. 
>>> b = np.array([1, 2, 3]) 
>>> a + b 
array([ 1.2, 2.2, 3.2]) 
>>> np.dot(a, ...
Indexing numpy array 
>>> a = np.array([[1, 2, 3], [4, 5, 6]]) 
>>> a[1, 2] 
6 
>>> a[1] 
array([4, 5, 6]) 
>>> a[:, 2] 
a...
Reference and documentation 
I NumPy User Guide: 
http://docs.scipy.org/doc/numpy/user/ 
I NumPy Reference: 
http://docs.s...
26 / 37
scikit-learn Machine Learning in Python 
I Simple and efficient tools for data mining and data analysis 
I Accessible to e...
A bug or need help? 
I Mailing-list: 
scikit-learn-general@lists.sourceforge.net; 
I Tag scikit-learn on Stack Overflow. 
...
Digits classification task 
# Load some data 
from sklearn.datasets import load_digits 
digits = load_digits() 
X, y = dig...
Data exploration and visualization 
# Data visualization 
import matplotlib.pyplot as plt 
plt.gray() 
plt.matshow(digits....
Fit a supervised learning model 
from sklearn.svm import SVC 
clf = SVC() # Instantiate a classifier 
# API The base objec...
Cross validation 
from sklearn.svm import SVC 
from sklearn.cross_validation import KFold 
scores = [] 
for train, test in...
Hyper-parameter optimization 
from sklearn.svm import SVC 
from sklearn.cross_validation import cross_val_score 
parameter...
Visualizing hyper-parameter search 
import matplotlib.pyplot as plt 
plt.figure() 
plt.plot(parameters, scores) 
plt.xlabe...
Estimator cooking: transformer union and pipeline 
from sklearn.preprocessing import StandardScaler 
from sklearn.pipeline...
Model persistence 
from sklearn.externals import joblib 
# Save the model for later 
joblib.dump(clf, "model.joblib") 
# L...
Reference and documentation 
I User Guide: 
http://scikit-learn.org/stable/user_guide.html 
I Reference: http: 
//scikit-l...
Upcoming SlideShare
Loading in …5
×
Upcoming SlideShare
Machine Learning with scikit-learn
Next
Download to read offline and view in fullscreen.

8

Share

Download to read offline

Numerical tour in the Python eco-system: Python, NumPy, scikit-learn

Download to read offline

We first present the Python programming language and the NumPy package for scientific computing. Then, we devise a digit recognition system highlighting the scikit-learn package.

Related Books

Free with a 30 day trial from Scribd

See all

Related Audiobooks

Free with a 30 day trial from Scribd

See all

Numerical tour in the Python eco-system: Python, NumPy, scikit-learn

  1. 1. Numerical tour in the Python eco-system Python, NumPy, scikit-learn Arnaud Joly October 2, 2014
  2. 2. 2 / 37
  3. 3. How to install Python? Download and use the Anaconda python distribution https://store.continuum.io/cshop/anaconda/. It comes with all the scientific python stack. Alternatives: linux packages, pythonxy, canopy, . . . 3 / 37
  4. 4. Using the python interpreter Interactive mode 1. Start a python shell $ ipython 2. Write python code >>> print("Hello World!") Hello World! Script mode 1. hello.py print("Hello World!") 2. Launch the script $ ipython hello.py Hello world! 4 / 37
  5. 5. Basic types Integer >>> 5 5 >>> a = 5 >>> a 5 Float >>> pi = 3.14 complex >>> c = 1 - 1j boolean >>> b = 5 > 3 # 5 <= 3 >>> b True # False string >>> s = ’hello!’ # Also works with "hello!" >>> s ’hello !’ 5 / 37
  6. 6. Python is a dynamic program language Variable types are implicitly inferred during the assignment. Variables are not declared. >>> # In python >>> a = 1 By contrast in statically typed language, you must declared the type. // In java, c, c++ int a = 1 6 / 37
  7. 7. Numbers and their arithmetic operations (+,-,/,//,*,**,%) >>> 1 + 2 4 >>> 50 - 5 * 6 20 >>> 2 / 3 # with py3 0.66... 0 >>> 2. / 3 # float division in py2 and py3 0.6666666666666666 >>> 4 // 3 # Integer division with py2 and py3 1 >>> 5 ** 3.5 # exponent 279.5084971874737 >>> 4 % 2 # modulo operation 0 7 / 37
  8. 8. Playing with strings >>> s = ’Great day!’ >>> s ’Great day!’ >>> s[0] # strings are sequences ’G’ >>> """A very very long string """ ’A verynvery long stringn’ >>> ’i={0} f={2} s={1}’.format(1, ’test’, 3.14) ’i=1 f=3.14 s=test’ 8 / 37
  9. 9. list, an ordered collection of objects Instantiation >>> l = [] # an empty list >>> l = [’spam’, ’egg’, [’another list’], 42] Indexing >>> l[1] ’egg’ >>> l[-1] # n_elements - 1 42 >>> l[1:2] # a slice ["egg", [’another list’]] Methods >>> len(l) 4 >>> l.pop(0) ’spam’ >>> l.append(3) >>> l [’egg’, [’another list’], 42, 3] 9 / 37
  10. 10. dict, an unordered and associative data structure of key-value pairs Instantiation >>> d = {1: "a", "b": 2, 0: [4, 5, 6]} >>> d {0: [4, 5, 6], 1: ’a’, ’b’: 2} Indexing >>> d[’b’] 2 >>> ’b’ in d True Insertion >>> d[’new’] = 56 >>> d {0: [4, 5, 6], 1: ’a’, ’b’: 2, ’new’: 56} Deletion >>> del d[’new’] >>> d {0: [4, 5, 6], 1: ’a’, ’b’: 2} 10 / 37
  11. 11. dict, an unordered and associative data structure of key-value pairs Methods >>> len(d) 3 >>> d.keys() [0, 1, ’b’] >>> d.values() [[4, 5, 6], ’a’, 2] 11 / 37
  12. 12. Control flow: if / elif / else >>> x = 3 >>> if x == 0: ... print("zero") ... elif x == 1: ... print("one") ... else: ... print("A big number") ... ’A big number’ Each indentation level corresponds to a block of code 12 / 37
  13. 13. Control flow: for loop >>> l = [0, 1, 2, 3] >>> for a in l: # Iterate over a sequence ... print(a ** 2) 0 1 4 Iterating over sequence of numbers is easy with the range built-in. >>> range(3) [0, 1, 2] >>> range(3, 10, 3) [3, 6, 9] 13 / 37
  14. 14. Control flow: while >>> a, b = 0, 1 >>> while b < 50: # while True do ... ... a, b = b, a + b ... print(a) ... 1 1 2 3 5 8 13 21 34 14 / 37
  15. 15. Control flow: functions >>> def f(x, e=2): ... return x ** e ... >>> f(3) 9 >>> f(5, 3) 125 >>> f(5, e=3) 125 Function arguments are passed by reference in python. Be aware of side effects: mutable default parameters, inplace modifications of the arguments. 15 / 37
  16. 16. Classes and object >>> class Counter: ... def __init__(self, initial_value=0): ... self.value = initial_value ... def inc(self): ... self.value += 1 ... >>> c = Counter() # Instantiate a counter object >>> c.value # Access to an attribute 0 >>> c.inc() # Call a method >>> c.value 1 16 / 37
  17. 17. Import a package >>> import math >>> math.log(3) 1.0986122886681098 >>> from math import log >>> log(4) 1.3862943611198906 You can try "import this" and "import antigravity". 17 / 37
  18. 18. Python reference and tutorial I Python Tutorial : http://docs.python.org/tutorial/ I Python Reference : https://docs.python.org/library/ How to use the "?" in ipython? In [0]: d = {"a": 1} In [1]: d? Type: dict String Form:{’a’: 1} Length: 1 Docstring: dict() -> new empty dictionary dict(mapping) -> new dictionary initialized from a mapping object’s (key, value) pairs dict(iterable) -> new dictionary initialized as if via: d = {} for k, v in iterable: d[k] = v dict(**kwargs) -> new dictionary initialized with the name=value pairs in the keyword argument list. For example: dict(one=1, two=2) 18 / 37
  19. 19. 19 / 37
  20. 20. NumPy NumPy is the fundamental package for scientific computing with Python. It contains among other things: I a powerful N-dimensional array object, I sophisticated (broadcasting) functions, I tools for integrating C/C++ and Fortran code, I useful linear algebra, Fourier transform, and random number capabilities With SciPy, it’s a replacement for MATLAB(c). 20 / 37
  21. 21. 1-D numpy arrays Let’s import the package. >>> import numpy as np Let’s create a 1-dimensional array. >>> a = np.array([0, 1, 2, 3]) >>> a array([0, 1, 2, 3]) >>> a.ndim 1 >>> a.shape (4,) 21 / 37
  22. 22. 2-D numpy arrays Let’s import the package. >>> import numpy as np Let’s create a 2-dimensional array. >>> b = np.array([[0, 1, 2], [3, 4, 5]]) >>> b array([[ 0, 1, 2], [ 3, 4, 5]]) >>> b.ndim 2 >>> b.shape (2, 3) Routine to create array: np.ones, np.zeros,. . . 22 / 37
  23. 23. Array operations >>> a = np.ones(3) / 5. >>> b = np.array([1, 2, 3]) >>> a + b array([ 1.2, 2.2, 3.2]) >>> np.dot(a, b) 1.200000 >>> ... Many functions to operate efficiently on arrays : np.max, np.min, np.mean, np.unique, . . . 23 / 37
  24. 24. Indexing numpy array >>> a = np.array([[1, 2, 3], [4, 5, 6]]) >>> a[1, 2] 6 >>> a[1] array([4, 5, 6]) >>> a[:, 2] array([3, 6]) >>> a[:, 1:3] array([[2, 3], [5, 6]]) >>> b = a > 2 >>> b array([[False, False, True], [ True, True, True]], dtype=bool) >>> a[b] array([3, 4, 5, 6]) 24 / 37
  25. 25. Reference and documentation I NumPy User Guide: http://docs.scipy.org/doc/numpy/user/ I NumPy Reference: http://docs.scipy.org/doc/numpy/reference/ I MATLAB to NumPy: http://wiki.scipy.org/NumPy_for_Matlab_Users 25 / 37
  26. 26. 26 / 37
  27. 27. scikit-learn Machine Learning in Python I Simple and efficient tools for data mining and data analysis I Accessible to everybody, and reusable in various contexts I Built on NumPy, SciPy, and matplotlib I Open source, commercially usable - BSD license 27 / 37
  28. 28. A bug or need help? I Mailing-list: scikit-learn-general@lists.sourceforge.net; I Tag scikit-learn on Stack Overflow. How to install? I It’s shipped with Anaconda. I http://scikit-learn.org/stable/install.html 28 / 37
  29. 29. Digits classification task # Load some data from sklearn.datasets import load_digits digits = load_digits() X, y = digits.data, digits.target How can we build a system to classify images? What is the first step? 29 / 37
  30. 30. Data exploration and visualization # Data visualization import matplotlib.pyplot as plt plt.gray() plt.matshow(digits.images[0]) plt.show() What else can be done? 30 / 37
  31. 31. Fit a supervised learning model from sklearn.svm import SVC clf = SVC() # Instantiate a classifier # API The base object, implements a fit method to learn from clf.fit(X, y) # Fit a classifier with the learning samples # API Exploit the fitted model to make prediction clf.predict(X) # API Get a goodness of fit given data (X, y) clf.score(X, y) # accuracy=1. What do you think about this score of 1.? 31 / 37
  32. 32. Cross validation from sklearn.svm import SVC from sklearn.cross_validation import KFold scores = [] for train, test in KFold(len(X), n_folds=5, shuffle=True): X_train, y_train = X[train], y[train] X_test, y_test = X[test], y[test] clf = SVC() clf.fit(X_train, y_train) scores.append(clf.score(X_test, y_test)) print(np.mean(scores)) # 0.44... ! What do you think about this score of 0.44? Tip: This could be simplified using the cross_val_score function. 32 / 37
  33. 33. Hyper-parameter optimization from sklearn.svm import SVC from sklearn.cross_validation import cross_val_score parameters = np.linspace(0.0001, 0.01, num=10) scores = [] for value in parameters: clf = SVC(gamma=value) s = cross_val_score(clf, X, y=y, cv=5) scores.append(np.mean(s, axis=0)) print(np.max(scores)) # 0.97... ! Tip: This could be simplified using the GridSearchCV meta-estimator. 33 / 37
  34. 34. Visualizing hyper-parameter search import matplotlib.pyplot as plt plt.figure() plt.plot(parameters, scores) plt.xlabel("Gamma") plt.ylabel("Accuracy") plt.savefig("images/grid.png") 34 / 37
  35. 35. Estimator cooking: transformer union and pipeline from sklearn.preprocessing import StandardScaler from sklearn.pipeline import make_pipeline # API Transformer has a transform method clf = make_pipeline(StandardScaler(), # More transformers here SVC()) from sklearn.pipeline import make_union from sklearn.preprocessing import PolynomialFeatures union_transformers = make_union(StandardScaler(), # More transformers here PolynomialFeatures()) clf = make_pipeline(union_transformers, SVC()) 35 / 37
  36. 36. Model persistence from sklearn.externals import joblib # Save the model for later joblib.dump(clf, "model.joblib") # Load the model clf = joblib.load("model.joblib") 36 / 37
  37. 37. Reference and documentation I User Guide: http://scikit-learn.org/stable/user_guide.html I Reference: http: //scikit-learn.org/stable/modules/classes.html I Examples: http: //scikit-learn.org/stable/auto_examples/index.html 37 / 37
  • shanugandhi7

    Feb. 3, 2020
  • SuguangDou

    Jan. 4, 2018
  • brussbowman

    Nov. 15, 2015
  • choeungjin

    Sep. 13, 2015
  • dragon515

    Mar. 11, 2015
  • johanparotte

    Oct. 15, 2014
  • JashenthreeGovender

    Oct. 4, 2014
  • vchollati

    Oct. 3, 2014

We first present the Python programming language and the NumPy package for scientific computing. Then, we devise a digit recognition system highlighting the scikit-learn package.

Views

Total views

3,112

On Slideshare

0

From embeds

0

Number of embeds

51

Actions

Downloads

64

Shares

0

Comments

0

Likes

8

×