1
00:00:00,570 --> 00:00:00,990
Hello.

2
00:00:01,020 --> 00:00:02,370
Welcome back.

3
00:00:02,370 --> 00:00:09,960
There are two popularly used sets of methods for describing convolution or layers in the first method

4
00:00:10,050 --> 00:00:12,470
which we have over here on the left.

5
00:00:12,630 --> 00:00:19,840
The constitutional net is viewed as a small number of relatively complex layers with each layer have

6
00:00:19,830 --> 00:00:21,850
in many stages.

7
00:00:22,050 --> 00:00:30,480
In this method there is a one to one mapping between can 0 10 source and network layers in the second

8
00:00:30,480 --> 00:00:32,070
method which is over here

9
00:00:35,850 --> 00:00:36,600
on the right.

10
00:00:36,630 --> 00:00:44,700
As you can see the convolution on your network is viewed as a large number of simple layers every step

11
00:00:44,790 --> 00:00:49,740
of process is regarded as a layer in its own right.

12
00:00:49,740 --> 00:00:56,490
This means that not every layer has parameters and we shall see the differences between these two as

13
00:00:56,490 --> 00:00:59,140
we dive deeper in the course.

14
00:00:59,250 --> 00:01:03,090
So that's all there is for this lesson and I shall see you in the next lesson.