Data Science, what even?!
2 likes1,846 views
The document discusses the fundamentals of data science, emphasizing the scientific method and the importance of hacking skills, statistical knowledge, and expertise. It highlights tools and programming languages like R and Python used for data analysis, modeling, and visualization. The author encourages a focus on relevant data and actionable insights to facilitate meaningful discussions and introspection in data-driven projects.
![map('state', region = c(row.names(USArrests)), col=cm.colors(16, 1)[floor(USArrests$Rape/max(USArrests$Rape)*28)], fill=T)](https://image.slidesharecdn.com/datasciencewebsummit-131101141308-phpapp02/85/Data-Science-what-even-89-320.jpg)
![from scipy.stats import
describe
s = [1,2,1,3,4,5]
print describe(s)](https://image.slidesharecdn.com/datasciencewebsummit-131101141308-phpapp02/85/Data-Science-what-even-94-320.jpg)
![import scipy.stats.poisson
p = poisson.pmf([1,2,3,4,1,2,3], 2)](https://image.slidesharecdn.com/datasciencewebsummit-131101141308-phpapp02/85/Data-Science-what-even-98-320.jpg)
![import numpy as np
x = np.array([ [1, 0], [0, 1] ])
vec, val = np.linalg.eig(x)
np.linalg.eigvals(x)](https://image.slidesharecdn.com/datasciencewebsummit-131101141308-phpapp02/85/Data-Science-what-even-103-320.jpg)
![>>> np.linalg.eig(x)
(
array([ 1., 1.]),
array([
[ 1., 0.],
[ 0., 1.]
])
)](https://image.slidesharecdn.com/datasciencewebsummit-131101141308-phpapp02/85/Data-Science-what-even-104-320.jpg)
![from sklearn import tree
X = [[0, 0], [1, 1]]
Y = [0, 1]
clf = tree.DecisionTreeClassifier()
clf = clf.fit(X, Y)
clf.predict([[2., 2.]])
>>> array([1])](https://image.slidesharecdn.com/datasciencewebsummit-131101141308-phpapp02/85/Data-Science-what-even-109-320.jpg)
![from pandas import *
x = DataFrame([
{"age": 26},
{"age": 19},
{"age": 21},
{"age": 18}
])
print x[x['age'] > 20].count()
print x[x['age'] > 20].mean()](https://image.slidesharecdn.com/datasciencewebsummit-131101141308-phpapp02/85/Data-Science-what-even-116-320.jpg)
Data Science, what even?!
- 5. And I cook.. A lot.
- 6. (n-1) items
- 7. Adapting.
- 8. Feedback.
- 11. Young mathematically inclined minds We knew everything.
- 13. So we asked “experts”.
- 15. Bad Data
- 16. Tasted like sh*t
- 20. From Our Results We had questions.
- 23. Find a Question Your Hypothesis
- 24. Current Data What do you have?
- 25. Features & Tests Try it.
- 26. Analyse Results Won’t be pretty.
- 28. But....
- 29. Good Discussions Imply good data scientists
- 30. Hacking Skills
- 36. Business Don’t need an MBA
- 37. In other words.
- 38. 1. Hacking 2. Maths & Stats 3. Expertise
- 40. 1. Question 2. Current Data 3. Features/Tests 4. Analyse 5. Converse
- 41. Find a Question Let’s imagine Github
- 42. Upgrade Repos Affect users as little as possible
- 43. import csv content = csv.read('repo1.csv')
- 44. λ e f (k; λ ) = k! k −k for k >= 0
- 48. KPIs are key Indicators from experience
- 50. Just test it..
- 54. We are Human. Emotional Connection
- 57. Focus on Data Relevant to your KPIs.
- 59. Data gives you the what Humans give you the why
- 60. Turn Information
- 63. Seeing, Feeling it The brain sees.
- 64. Not regressions
- 65. Not p-values
- 66. Not slopes
- 67. Not F-statistics
- 68. Not coefficients
- 72. Toolbox What do we use?
- 74. RStudio The IDE
- 75. lubridate and zoo Dealing with Dates...
- 79. RCurl, RJSONIO Find more Data
- 88. maptools Read & View Maps
- 89. map('state', region = c(row.names(USArrests)), col=cm.colors(16, 1)[floor(USArrests$Rape/max(USArrests$Rape)*28)], fill=T)
- 92. scipy.stats
- 94. from scipy.stats import describe s = [1,2,1,3,4,5] print describe(s)
- 97. λ e f (k; λ ) = k! k −k for k >= 0
- 98. import scipy.stats.poisson p = poisson.pmf([1,2,3,4,1,2,3], 2)
- 101. NumPy Linear Algebra
- 102. ⎛ 1 0 ⎞ ⎜ 0 1 ⎟ ⎝ ⎠
- 103. import numpy as np x = np.array([ [1, 0], [0, 1] ]) vec, val = np.linalg.eig(x) np.linalg.eigvals(x)
- 104. >>> np.linalg.eig(x) ( array([ 1., 1.]), array([ [ 1., 0.], [ 0., 1.] ]) )
- 107. NLTK Natural Language Tool Kit
- 109. from sklearn import tree X = [[0, 0], [1, 1]] Y = [0, 1] clf = tree.DecisionTreeClassifier() clf = clf.fit(X, Y) clf.predict([[2., 2.]]) >>> array([1])
- 112. Pattern Web Mining for Python
- 116. from pandas import * x = DataFrame([ {"age": 26}, {"age": 19}, {"age": 21}, {"age": 18} ]) print x[x['age'] > 20].count() print x[x['age'] > 20].mean()
- 117. Python vs R? Different Purposes
- 119. Original Question What is Data Science?
- 120. Back to you For questioning