summaryrefslogtreecommitdiff
path: root/04-08.md
blob: a39e9e53aab4ea7a5a518df98f8b9ccf54a711ba (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
## Outline
- Concept
	- Data Structure
	- Code Efficiency
- Big-O

## What is Data Structure?

- How data is organized in memory (or on disk)

|Data|Structure|
|----|---------|
|Atomic or composite data|A set of rules that holds the data together|

- Examples:
	- Linked List
	- Array

## Why do we care about Data Structure?

- Efficiency can refer to either 
	- Space (importance depending on architecture)
	- or Time
- We will focus on **time**, So, we want our operations efficient in terms of time

## How to measure efficiency?

- Wall-clock time?
	- Not so great
		- Not that accurate
		- can't compare across computer or programmers
		- dependent on circumstances and what's running

- Count the number of operations!

### What kind of Operations?

- insert (add)
- delete (remove)
- access/retrieval
- min/max
- union/merge (less common)

## A Simple Example

- We assume each basic operation takes one step
- Example: Initialize an array (a loop setting `a[i]=0`)
- Count the steps ... how many times is ...

Using a for loop:
```
for(i = 0; i<n; i++){
	a[i] = 0;
}
```
- Break down the steps:
	1. `i = 0` (done once)
	2. `i < n` (done 'n+1' times)
	3. `a[i] = 0` (done 'n' times)
	4. `i++` (done 'n' times)
- The total number of steps is `1 + (n+1) + n + n` = `3n+2`

- This is tedious and error prone, so we will teach how to count and be lazy!

## When do we care?

- Turns out we care about what happens **when n is big**, i.e., how our alogirthm behaves as n gets large

- When n is a million, 2 doesn't matter, 2000 doesn't matter, c doesn't matter for sufficiently larger n

- We only care about the one with the highest impact on efficiency

## Big-O Notation

- To classify the cost of an algorithm we use the big-O notation
	- it's called that because we use a capital 'O'

- To compute the big-O runtime of a function (mathematical function):
	- drop all but the fastest-growing term
	- drop any coefficient on that remaining term
- So, e.g., `3n+2` is O(n)
	- Drop the 2 (not the fastest growing term) and the 3 (coefficient of remaining term)

- Example: `5n^2 + 3n + 5` -> `5n^2` -> `n^2`
	- Would be represented as O(n^2)

- Example: `600 + 30n` -> `30n` -> `n`
	- Would be represented as O(n)

- Example: `15n + 12nlog(n) + 3` -> `12nlog(n)` -> `nlog(n)`
	- Would be represented as O(nlog(n))

- Example: `14n^2 + 0.05n^3 + 18000000` -> `0.05n^3` -> `n^3`
	- Would be represented as O(n^3)

## Apply Big-O on Code Analysis

- Let's analyze some loops and compute the big-O runtimes
	- `for(i=0; i<n; i++) x++;`
		- `3n+2` -> `n` -> O(n)
	- `for(i=n; i>0; i--) x++;`
		- O(n)
	- `for(i=0; i<(n/2); i++) x++;`
		- cut the threshold by half
		- O(n/2) is **not** the finalized runtime
		- O(n) is the finalized runtime
	- `for(i=0; i<n; i=i+2) x++;`
		- Double the step size
		- O(n/2) is **not** the finalized runtime
		- O(n) is the finalized runtime

## Big-O Add

- O(n) + O(n) = ?
	- `for(i=0; i<n; i++) x++; O(n)`
	- `for(i=0; i<n; i++) x++; O(n)`
	- O(n) + O(n) = O(n + n) = O(2n) -> O(n)

- `for(i = 1; i < n; i=i*2) x++`
	- Break it down:
		- iteration 1: i=1
		- iteration 2: i=2
		- iteration 3: i=4
		- iteration 4: i=8
		- iteration 5: i=16
		- and so on, until i<n
		- Notice: iteration x: i=2^(x-1)
	- The total number of iterations we need to run is x = (log2(n)+1)
		- O(x) = O(log2(n)+1) -> O(log2(n)) -> O(log(n))
			- log base value doesn't matter

- Example: `for(i=1; i<n; i=i*100) x++;`
	- O(log100(n)) -> O(log(n))
- Example: `for(i=1; i<n; i=i/5) x++;`
	- O(log(n))

Example: Nested For Loop

```
for(i=0; i<n; i++){
	for(j=0; j<n; j++){
		x++;
	}
}
```

- How many times does this need to run?

|iteration number|outer loop|inner loop|
|----------------|----------|----------|
|1               |i=0       |n         |
|2               |i=1       |n         |
|3               |i=2       |n         |
|n               |i=n-1     |n         |

- You get O(n * n) = O(n^2)
	- Essentially, you multiply a big-O (outer loop) by another big-O (inner loop)

Example: Another Nested For Loop

```
for(i=1; i<=n; i++){
	for(j=1; j<1; j++){
		x++;
	}
}
```

- These loops are no longer independent

|iteration number|outer loop|inner loop|
|----------------|----------|----------|
|1               |i=1       |0         |
|2               |i=2       |1         |
|3               |i=3       |2         |
|n               |i=n       |n-1       |

- Total number of iterations is x = 0+1+2+...+(n-1)
- What is x?
	- x = (n-1)+(n-2)+...+2+1+0
	- 2x = (n-1)\*n -> x = ((n-1)\*n)/2
- O(x) = O(((n-1)\*n)/2) -> O(n^2)

---

[04/10 ->](04-10.md)