1
00:00:00,000 --> 00:00:01,950
Welcome to the final week of

2
00:00:01,950 --> 00:00:04,860
this final course of
this specialization.

3
00:00:04,860 --> 00:00:06,630
There's a lot happening
this week that

4
00:00:06,630 --> 00:00:08,280
I hope you find fun and exciting.

5
00:00:08,280 --> 00:00:11,055
You get to take sunspot data,

6
00:00:11,055 --> 00:00:15,810
the real data and apply
a 1D convnet to that.

7
00:00:15,810 --> 00:00:18,285
So this is generalization
of the 2D curve you see,

8
00:00:18,285 --> 00:00:20,100
and on top of that and LCM and on

9
00:00:20,100 --> 00:00:21,890
top of that it density
or networks.

10
00:00:21,890 --> 00:00:24,360
So a lot of pieces you've
learned will come together

11
00:00:24,360 --> 00:00:26,940
to help you build
an effective sunspot prediction.

12
00:00:26,940 --> 00:00:29,535
This specialization was called
Tensor Flow in practice.

13
00:00:29,535 --> 00:00:31,215
So one of the things that

14
00:00:31,215 --> 00:00:33,180
Andrew and I have been
really excited about

15
00:00:33,180 --> 00:00:35,580
is helping you to be able to put

16
00:00:35,580 --> 00:00:36,900
everything that you've
been learning in

17
00:00:36,900 --> 00:00:39,125
this specialization
into practice.

18
00:00:39,125 --> 00:00:40,805
So in this course
we've been using

19
00:00:40,805 --> 00:00:43,100
synthetic data for
the last three weeks,

20
00:00:43,100 --> 00:00:45,710
but we're now taking you away
from the synthetic data and

21
00:00:45,710 --> 00:00:47,420
giving you the
real-world sunspot data

22
00:00:47,420 --> 00:00:48,740
that Andrew mentioned;

23
00:00:48,740 --> 00:00:51,585
convolutional neural
networks and 1D cams,

24
00:00:51,585 --> 00:00:53,340
LSTMs, deep neural networks,

25
00:00:53,340 --> 00:00:54,650
we'll be able to mash them

26
00:00:54,650 --> 00:00:56,030
all together and
hopefully come up

27
00:00:56,030 --> 00:00:58,915
with some nice prediction
engines for sunspot activity.

28
00:00:58,915 --> 00:01:00,140
So by the end of this week,

29
00:01:00,140 --> 00:01:01,160
you have built a pretty

30
00:01:01,160 --> 00:01:03,080
sophisticated neural
network model

31
00:01:03,080 --> 00:01:05,255
that it puts together
many of these pieces.

32
00:01:05,255 --> 00:01:07,595
By doing that, I hope you may be

33
00:01:07,595 --> 00:01:09,815
able to start seeing
how you could

34
00:01:09,815 --> 00:01:12,170
maybe assemble these pieces
you've learned in

35
00:01:12,170 --> 00:01:15,830
Tensor Flow to build other
rich neural networks as well.

36
00:01:15,830 --> 00:01:18,570
So please go on to
the next video.