00:13Welcome back, and peace be upon you. Now we'll continue with the course introduction.
00:18We continued with machine learning and are now going through the necessary steps to make
00:25The machine learning model we're following works correctly and gives us
00:33The third step is vision engineering. Now you're wondering
00:38What are features? Features are tools, or more accurately, input data.
00:44Which you input into your model to create a reproduction or
00:49This works for you, so here are the features. Now, features are in features engineering.
00:54From its name, what can you do? New features from old features, meaning you
01:00You have these features or input data; you take some of them and use them to create something.
01:05New features. Or, for example, you have many features that you can choose from.
01:11These are features. Why do we do features engineering? We do features.
01:18Engineering for what? We enhance the data to make it work.
01:23What's a representation better for? For problems we can normally work on. So that's it.
01:29We will discuss features engineering in detail during this course. The step
01:36The fourth is choosing the aming algorith Bengal. The choice is, is it in?
01:41We have many machine learning algorithms. Now, how do we choose an algorithm?
01:47The appropriate model for our situation. What do we need to look at in terms of data? Now, data.
01:54It's what decides which algorithm you want to choose. Now we
02:00It has, in our machine, three catacrezes. The first catacreze
02:04Which is Supervised Learning. Supervised Learning includes
02:08There are two types of algorithms. The first type is the algorithm.
02:12Regression. And the other algorithms, which are
02:14Classification. Now, it's the most important thing in Supervised Learning.
02:20Data. Data in Supervised Learning
02:24It will be labeled data. What does that mean?
02:27Label data. I have Y = ix + b. Now, Y is
02:33The output is X + B. Now, Y is the one I want.
02:39I used the model to create it, where the Y and X are the features.
02:46The features I want to add to this model so that it works for me
02:49Prediction for Y. Supervision learning must have output.
02:54Because this is the output I want to create
02:56I use models to do
02:59Of course, using the X
03:02The X represents the features of the embedded data.
03:05We will, of course, discuss the A and B in detail later.
03:08Okay, so we said that we have
03:11Oh, or two types
03:15Who?
03:17Two types of algorithms
03:19Which is in Supervision Learning
03:21In our algorithm, the regeneration
03:24And in our classification
03:25Now, the viGriReshan algorithm is what we use.
03:28If we have the diatia
03:29or output
03:31Follow our Diata
03:33It means the Y was continuous
03:35What does Continius mean?
03:36So you want to do
03:48Use the regeneration
03:49When you are
03:51The-Y
03:52The-Output
03:53We have
03:53Continuous
03:54while
03:55Classification
03:55It will be
03:57The-Y
03:57Classes
03:58For example
03:59I have
04:00The-Output
04:00Follow the data
04:02Follow me
04:02Yes or No
04:03False or True
04:05Or for example
04:06Multiple Classes
04:07I mean, I have
04:07More than
04:08for example
04:09I have pictures
04:09I want to do it for her
04:11Classified
04:12For example
04:13I want to leave
04:14computer
04:14Follow me
04:15He knows
04:16What is this picture?
04:17Hi
04:18It is
04:19Classification
04:20or Multiple Classification
04:22Now it will become clear
04:23The picture is more
04:24Bal-Example
04:25Let me show you
04:26how
04:26Hello, for example, me
04:27I have this data
04:28Hi data
04:29It contains pictures
04:30what
04:31pictures of animals
04:32So I have
04:33For example, here
04:34horse
04:35I have beauty
04:36I have
04:36dog
04:37I have
04:38cat
04:38or a cat
04:39Hello, I want to leave
04:41Enter this data
04:42on the computer
04:44I'll make him work for me
04:45what
04:45It appears to me
04:46Output
04:47He tells me
04:47This is a horse.
04:48Hadi is a cat
04:49Hadi Jamal
04:50Hadi Kalb
04:50This
04:51This, of course
04:53Supervised Learning
04:54Because I have
04:55It will be
04:57Output
04:58Which is
04:59The-Y
04:59I want to do it
05:00Prediction
05:01Production
05:02via
05:02I'm going in
05:03image
05:04for example
05:04I'm inserting a picture
05:05for example
05:06Fortification
05:07And then
05:08what
05:08I'm going in
05:10This immunity
05:10This picture
05:11World of the model
05:12Fbiji
05:12what
05:13Model
05:13He does it for me
05:13Prediction
05:14She is Hadi
05:15image
05:15Fortification
05:16What do I do? Where do I input this data into my model? My model does...
05:22I entered this data into a class, and what result did I get? The result I got was that this is the image.
05:27This picture is of a horse, this picture is of a camel, this picture is of a cat, and this picture is of a dog. So this is it.
05:34This means that I'm doing this and making the computer recognize me when I enter an image.
05:41Picture of any animal. This is Supervision Learning. It's in
05:47I have it here. I entered the data (the input) into the model.
05:52I got an output, which is the Y-axis. This is Supervision Learning. Meanwhile
05:58In the second category, which is now Supervision Learning. Now
06:02Supervision Learning's most important feature is what? The data won't be in it.
06:10And what is this? Oh, I'll use the algorithms that make it for us.
06:17Identify the pattern, or does it mean you're doing a clustering? No, for example, I entered
06:23It has pictures, for example, of similar images. Let me show you how. For example, me
06:28I have the same data that contains pictures of the animals. What should I do?
06:33I enter it into my model. And then what happens to the model? It does something for me.
06:37Clusting means taking, for example, horses. That's for example, me.
06:43And I put them in a group. He takes them for me, but he doesn't know, for example, he doesn't tell me.
06:47These are horses. No, this one takes them because they have things in them.
06:53Similar. Or he puts them together. For example, he creates clusters.
06:57For beauty. It creates clusters for cats and dogs. So it puts them in groups.
07:04These groups contain similar images. This is now
07:08Supervise Learning doesn't have label data. What does clustering do?
07:13Grouping similar images or tools. While Supervision Learning, what is it? It is
07:22Ah, view label data. That means there's output. So you need to make the computer...
07:28He works for you, telling me, for example, "This is a picture of a horse." Okay. Now, here we will...
07:36We'll stop here, and God willing, we'll continue with the remaining steps in the next video. Thank you.
07:41I'll see you in the next video.
07:52music
