1
00:00:00,000 --> 00:00:03,767
[MUSIC]

2
00:00:03,767 --> 00:00:06,625
We're not starting from
scratch in this course, and

3
00:00:06,625 --> 00:00:10,040
we're going to assume some
background from all of us.

4
00:00:10,040 --> 00:00:12,900
In fact, we're going to assume that
you've taken the first two courses in

5
00:00:12,900 --> 00:00:13,790
the specialization.

6
00:00:13,790 --> 00:00:15,510
This is really fundamental.

7
00:00:15,510 --> 00:00:16,890
From the first course,

8
00:00:16,890 --> 00:00:21,200
you gained a wide view of what's
possible to do with machine learning.

9
00:00:21,200 --> 00:00:23,410
And if you already have that view,
it's okay.

10
00:00:23,410 --> 00:00:26,140
But this is something that's going
to be very helpful, as well as

11
00:00:26,140 --> 00:00:30,570
facilitating kind of your programming
skills, manipulating data and all that.

12
00:00:30,570 --> 00:00:35,810
We're also going to be taking the second
course, where we cover issues like what's

13
00:00:35,810 --> 00:00:40,514
a machine learning algorithm, gradient
ascent, over-fitting, validation set and

14
00:00:40,514 --> 00:00:44,050
cross-validation, bias-variance trade off,
and regularization.

15
00:00:44,050 --> 00:00:46,140
So we'll see if we remember those topics.

16
00:00:46,140 --> 00:00:47,870
Go back and review if you don't.

17
00:00:47,870 --> 00:00:51,760
If you haven't taken any of the courses
but you know what the techniques are, feel

18
00:00:51,760 --> 00:00:55,980
free to jump straight into this course but
we'll assume them throughout the course.

19
00:00:58,230 --> 00:01:01,860
Just like the other courses we'll
assume some basic background so

20
00:01:01,860 --> 00:01:05,260
you're going to take a few
derivatives here and there.

21
00:01:05,260 --> 00:01:08,960
You should know what a vector is
because we're going to use those.

22
00:01:08,960 --> 00:01:10,900
And we're going to use
some basic functions,

23
00:01:10,900 --> 00:01:14,050
including exponentiation and logarithm.

24
00:01:14,050 --> 00:01:17,800
So if you need to take a little
refresher on those basic functions,

25
00:01:17,800 --> 00:01:18,880
this is a good time to do it.

26
00:01:19,930 --> 00:01:23,950
This is a hands-on course, for every
module you're going to do some programming

27
00:01:23,950 --> 00:01:27,320
with real world data and
get real world results.

28
00:01:27,320 --> 00:01:28,770
So you should expect that, but

29
00:01:28,770 --> 00:01:31,380
that doesn't mean that you're
going to have to do some programming.

30
00:01:31,380 --> 00:01:35,440
We've set up an infrastructure makes it
easy for you to do it if you know Python.

31
00:01:35,440 --> 00:01:39,480
If you don't know Python, you can catch up
on python pretty quickly just like you did

32
00:01:39,480 --> 00:01:44,450
in previous courses or you can implement
in whatever language you want.

33
00:01:44,450 --> 00:01:47,090
We don't assume that you use
a particular language, but

34
00:01:47,090 --> 00:01:49,840
it would be a lot easier
if you use Python.

35
00:01:49,840 --> 00:01:53,030
Unlike this first course in
the specialization we'll use heavy use of

36
00:01:53,030 --> 00:01:54,630
GraphLab Create.

37
00:01:54,630 --> 00:01:57,340
Because use a good black
box to get started.

38
00:01:57,340 --> 00:02:02,360
In this course, we're going to not
really heavily on GraphLab Create.

39
00:02:02,360 --> 00:02:04,210
We do suggest that you use SFrames,

40
00:02:04,210 --> 00:02:08,483
which is the data manipulation open source
library that was created by Dato, or

41
00:02:08,483 --> 00:02:13,340
you can use other libraries like
pandas if you prefer in Python.

42
00:02:13,340 --> 00:02:17,440
And there will be some assignments,
where using the pre-implemented algorithm,

43
00:02:17,440 --> 00:02:21,750
just try to understand how things behave
before you implement them yourself.

44
00:02:21,750 --> 00:02:23,790
For that,
we do suggest you use GraphLab Create, but

45
00:02:23,790 --> 00:02:26,620
you can use other libraries
like scikit-learn.

46
00:02:26,620 --> 00:02:27,810
It's up to you.

47
00:02:27,810 --> 00:02:33,200
It will be easier if you use Python,
and we will give you

48
00:02:33,200 --> 00:02:37,450
some starter code and some other things,
but you are welcome to use your own.

49
00:02:37,450 --> 00:02:40,090
But the net result of this course

50
00:02:40,090 --> 00:02:44,860
is all about you implementing your own
machinery algorithms from scratch.

51
00:02:44,860 --> 00:02:46,940
So that's what you should
be prepared to do.

52
00:02:46,940 --> 00:02:51,450
Now you will need a computer that
has a little bit of power to it.

53
00:02:51,450 --> 00:02:56,400
A laptop should be fine,
64-bit machine will make a big difference.

54
00:02:56,400 --> 00:02:59,620
You're going to need access to
the internet to download data sets and

55
00:02:59,620 --> 00:03:01,720
everything else, and
to watch these videos.

56
00:03:01,720 --> 00:03:05,670
And you have to have the ability to
install Python, maybe GraphLab Create if

57
00:03:05,670 --> 00:03:09,590
you choose to use it, and store a few
gigabytes of data on your machine.

58
00:03:09,590 --> 00:03:12,420
Just like you did with previous courses.

59
00:03:12,420 --> 00:03:14,980
We'll also provide some
other resources for

60
00:03:14,980 --> 00:03:16,640
those who don't have their own machines.

61
00:03:16,640 --> 00:03:21,240
They can use machines on the web, and
we'll talk about that too in our readings.

62
00:03:21,240 --> 00:03:25,399
[MUSIC]