See, here we have four topics, there are no problems..

All of them are four concepts, submission learning, deep learning and help is
generated..
Four concepts are now in the way of generating..
Generative is generated by two concepts, drag is generated..
If you do it very deep in these two models, it will be easier to add it..
Now I am going to give you a project idea..
You will be using the project idea that you have to work on in the ETIP..
I am going to tell you how to do this..
I am a classifier..
I am going to give you an input based on this..
So this should be in the class 2 or in the class 2..
This is your decision..
I have four other options, banking sector, my savings, my salary, my address, my
own house, rent and house Finally, I have to pay the loan to the bank, S or loan, S
is the loan and loan is the banking sector This is a healthcare partner..
My fever range, my cough range, my body condition, my weight, height, etc..
I don't have cancer..
Otherwise, cancer, heart beat, heart disease, etc..
All these are out..
This is also a classification..
Now, we will take a production company..
A Suzuki company..
Now in Suzuki company, if you add a car detail, it will be a bug or something, it
will be a car name..
So what type of setup, what type, what fuel is used, how many kms of mileage is
given, what is this, this bug..
So that is also a class..
Classification is the same concept..
You are going to predict this class and that class..
But every industry has its own use case..
Banking has its own use case and classification..
S is low-war..
So, whether you go to DBR or normal or current or salary, Or you can provide for
your home loan..
You can provide personal use and personal loan..
So, in classes like this, you need to think about the concept..
You need to think about the concept..
The data set you are going to give is there..
There are 10 preprocesses in that data set..
You can use 2-3 of those preprocesses..
But if you know those 10 preprocesses, you can easily select them..
You can select whether you are in the missing or the outlier..
You can change the number of categories in the course of the course..
So, there are 10 pre-processed steps in the course of the course..
Think about it..
If you have any project, you can do it..
So, you can do it in a healthcare format..
I decided to do a heart disease in cancer and put it in the resume..
What is your motive? What is the use case for building a project? I wanted to build
some use cases..
Why are you going to build a project? We have a website in the hospital..
We will predict the results if we give a detailed review of the fever..
Maybe the energy use also, provided use also, provided use also..
If we check everything, we can go with all types..
In the previous days, we can see if it is a heart disease or a heart disease, we
can take a test and wait for 14 days..
So now we will take that test for 1 day and then we will go with the test for 10
minutes and then we will predict what diseases we will have..
This is how each use case will be..
Now I will give you the use cases..
I will give you an image of the cat in the cat's eye..
So, what will you use in this? What will you use in the morning? I have an image, I
will load the image and the model will be in the drop-file pattern..
What is the model? Cn So in case, the image I am taking, I am going to give a
heart, an extra image, a soft copy of that, and I will give it to my model..
Now I want to get it in the body, in the canister..
So what is this? What is the classification of the model? Classification of the
model? Classification of the model? There is a photo image..
In that photo image, there is an X-ray, maybe a..
..
..
What is the classification of the image? How cute it is..
I will show you the video I will show you the video What is the model? Architecture
Architecture? CnR architecture This is image I will give you a video A pose is good
It is an element We will detect it We will use this I have a PDF document In that
PDF document I have the image, the image and the tables..
What I am going to do is, I will show you my personal document..
I will give you the document in detail..
I will show you the leak..
If you keep this, you can check the Q&A..
What model should I use? Rack model..
So, if you want to implement only the rack model, you can do the authentic model..
You can implement each and every one..
This is the base concept..
base concept..
This R27 is the base concept..
But the data set is the problem..
The problem is in your model..
Your model is going to perform the same..
But what you are going to do is, your data set will become pure..
So it will become pure..
If you are going to correct everything in a clean data, it will not work..
It will be done by the Tamil Focusing..
So you have to take time to collect the data..
Data is actually a pre-processed data..
Think about it..
There are 10 pre-processed data in machine learning..
There are 10 pre-processed data in deep learning..
There are 10 pre-processed data in NLV..
Similarly, for RAD, this is mixed in RAD..
RAD is generated by mix..
We will use this data to create something..
Think about it..
You can choose a healthcare company, a production company, an e-commerce website..
So if you have worked in an e-commerce website, you can choose a project that has a
classification..
In one project, you can choose one accuracy matrix, one model and one output..
Do you understand? You cannot influence everything..
You cannot inclement everything in one project..
If you develop a project, you can use only accuracy and classification..
If you add all the themes, you can use the converse..
Accuracy, of course, the week, precision, the chat GPT command, you can take the
exact key..
So, what is that? What is that? Logistic, decision tree, random parts, everything
you do, you can use..
When you ask the production, if you implement all these things, you will not be
able to do it..
This is one of the most important things in the production sector..
One project has become a domain visa..
Machine learning, deep learning, NLP and generative education..
So, in this, classification, machine learning is the main role of classification..
So, there is a very big concept in that..
Many conceptions are not very big..
You collect data from machine learning..
You check the data first, then the statistical view..
Then I will start the pre-processing..
In pre-processing, you have given a lot of steps to share an excel..
Models? There are no models..
What are the other steps? There are a lot of examples..
For example, in the three mission learning, pre-processing steps, you have put it
all in here, you have shared it with all of them..
This is a very difficult field..
But tricks or phase, tricks are the most important..
If I say it as a major issue, we are going to support you in a small way..
You are going to think logically in that support..
Okay, who has prepared the resume? Who has prepared the resume? Bro, all the
resumes are done, the project is over..
In that project, you are going to be classified as a machine learning..
In the generative way, there are three projects..
One is a ragging agenda model..
You can see that you have two projects in one project..
You can run a project and deploy it..
You can also deploy it in the same project..
So, your project will go to your residence and revalidate it..
After that, in one model, in one project, in one model, in one deployment, in one
accuracy matrix, in one point unit mechanisms..
But when you explain it, I will explain it..
How to explain it? I have chosen the final decision..
I have built and seen all the models..
When you look at all the models, when you check the train and test accuracy, the
train and test accuracy is good and the result is good..
I have given good results, so I took this final model and saved it and set it in
production..
So what is the basic question? Pre-processed steps are heard..
Pre-processed steps are heard..
Now let's see how you have logical thinking..
A small concept is that I am an Amazon company..
Many of my products are not available..
One day, an IR customer product is coming..
I will be sitting and answering your feedback..
I will be reading your feedback and then I will be able to acknowledge your
feedback..
You can acknowledge your feedback and then you can create your own account..
So what is the concept of implementing? The concept is that I am selling products,
I am sending a complaint message to customers, I want to do it immediately..
So what is the concept of implementing? How to do it? How to generate auto e-mail?
How to do it? It is very logical..
You are the one who is talking..
Think logically..
It is not a very big difference..
But we have to respond to the autograder..
We have to respond to the ebony..
You can do it in a chat..
You can do it in a chat..
How do you do it? You can do it in a HGP cake..
You can do it in a HGP cake..
You can do it in a classification..
What do you do before that? You have only mail You don't get anything in that mail
You don't get any complaint You don't get any instant mail You don't get any
feedback What do you do? You have to do sentiment analysis You have to do sentiment
analysis First, there is a mail There are 1000 mails You have to make it known to
your model If you get such a mail format You will know this problem What are you
going to do? You are going to correct it in 1000 miles..
You are going to send it in a cluster..
If you send it in a cluster, it will split the cluster..
What is it going to do? It will split the cluster in a cluster..
It will split it in a cluster and send it in classes..
If you go inside and read the classes in the cluster, you will know the class of
0..
You will know the name of the class..
Thank you for watching