1
00:00:00,510 --> 00:00:06,100
‫Now we have prepared our data for image classifier.

2
00:00:06,390 --> 00:00:10,350
‫The next step is model creation.

3
00:00:10,350 --> 00:00:15,330
‫There are two ways to create and create newer model indicators.

4
00:00:15,360 --> 00:00:21,510
‫The first one is the sequential model Apia and the second one is the functional API

5
00:00:24,320 --> 00:00:33,170
‫sequential API is straightforward and simple whereas functional API is little bit complex but it will

6
00:00:33,170 --> 00:00:43,330
‫give you the flexibility to create some complex neural networks sequential EPA is useful to create layer

7
00:00:43,330 --> 00:00:51,130
‫by layer models such as these where all the outputs of previous layer are connected as inputs of the

8
00:00:51,130 --> 00:00:53,710
‫next layer and so on.

9
00:00:53,770 --> 00:00:59,440
‫So for the simple neural networks like this sequential API is recommended

10
00:01:02,910 --> 00:01:12,030
‫but for some advanced complex structure such as this here we are using input as an input for concrete

11
00:01:12,030 --> 00:01:12,810
‫layer as well.

12
00:01:13,620 --> 00:01:22,860
‫So this layer have two inputs one all our primary input parameters and the outputs of hidden layer as

13
00:01:22,860 --> 00:01:25,700
‫well so anywhere.

14
00:01:25,710 --> 00:01:34,230
‫If you want complex structure like this or you want to only use some part of your input in one hidden

15
00:01:34,230 --> 00:01:41,070
‫layer and other part of input in some other hidden layer for all such variations you can use functional

16
00:01:41,090 --> 00:01:51,030
‫API but for a straight forward dense neural networks like this you can use sequential API in this course.

17
00:01:51,070 --> 00:01:57,660
‫First we will use the sequential API then we will also look at this example of functional labia.