Deep Learning using Keras – Complete & Compact Dummies Guide paid course free for all. You will Learn Computer Vision with CNN: Basic Python, Numpy, Pandas, Matplotlib, Keras Text MLP, VGGNet, ResNet, Custom Model in Colab

- Deep Learning
- Computer Vision
- Keras
- Machine Learning
- Python

### Deep Learning using Keras – Complete & Compact Dummies Guide Course Requirements

- Basic computer knowledge and an interest to learn the Deep Learning using Keras

## Deep Learning using Keras – Complete & Compact Dummies Guide Course Description

Welcome to my new course ‘Deep Learning from the Scratch using Python and Keras’.

As we all know, the field of artificial intelligence is roughly divided into deep learning and machine learning. In fact, deep learning itself is machine learning, but deep learning and its deep neural networks and algorithms try to learn advanced features from data without human intervention. This makes deep learning the foundation of all future automatic intelligent systems. In this course, I will start learning from very basic things, such as learning the basics of programming languages and other support libraries at the beginning and then turn to core topics. Take a look at the cool topics covered in this course.

First, we will have an introductory theoretical session on artificial intelligence, machine learning, and deep learning based on artificial neurons and neural networks. After that, we will prepare our computer for Python coding by downloading and installing the anaconda package. Check and see if everything is installed correctly. We will use a browser-based IDE called Jupyter notebook for additional coding exercises. I know that some of you may not have Python-based programming experience. The next few courses and examples will help you gain basic Python programming skills to continue the courses included in this course. Topics include Python mapping, flow control, functions and tuple lists, dictionaries, functions, etc.

Then we will start to learn the basics of the Python Numpy library, which is used to add support for large multi-dimensional arrays and arrays, as well as a large number of classes and functions. Next, we will learn the basics of the matplotlib library, which is a Python plotting library for corresponding numerical expressions in NumPy. Finally, there is the Pandas library, which is a software library for data manipulation and analysis written for the Python programming language. After the basics, we will install the deep learning libraries theano, TensorFlow, and API to handle these so-called Keras. We will write all future code in Keras. Then before entering deep learning, we will introduce in detail the basic structure of artificial neurons and how they are combined to form a theoretical course of artificial neural networks.

Then we will see what activation functions are, the different types of the most popular activation functions, and the different scenarios in which we must use them. Later we will learn about loss functions, different types of popular loss functions, and the different scenarios in which we must use them. Like the Trigger and Loss functions, we have an optimizer that can optimize the neural network based on training feedback. We will also see detailed information about the most popular optimizers and how to decide in which scenarios we must use them.

Then at the end, we will discuss the most popular types of deep learning neural networks and their basic structure and use cases. It is accurately divided into two halves. The first half is about deep learning multilayer neural network modeling for text-based data sets, and the second half is about building convolutional neural networks for image-based data sets. The first step of the King County housing price prediction model in the United States is to obtain the data set from the Kaggle website and upload it to our program, and then as the second step, we will perform EDA or analytical exploratory data analysis on the loaded data, and then we will prepare Data can be input into our deep learning model.

Then we will define the Keras deep learning model, once we define the model, we will compile the model, and then we will fit our data set to the compiled model and wait for the training to complete. After training, training history and metrics (such as accuracy, loss, etc.) can be evaluated and visualized using matplotlib. Finally, we have trained our model. We will try to use our deep learning model to predict real estate prices in Jinxian County and evaluate the results. That is a text-based regression model. We will now continue to use the text-based binary classification model. We will use a derived version of the heart disease dataset from the UCI machine learning repository. Our goal is to predict whether a person will have heart disease based on the learning obtained from this data set. Repeat the same steps here.

The first step is to get the data set and load it into our program. Then, as the second step, we will perform EDA or exploratory data analysis on the uploaded data, and then we will prepare the data for our deep learning model. Then we will define the Keras deep learning model, once we define the model, we will compile the model, and then we will fit our data set to the compiled model and wait for the training to complete. After training, training history and metrics (such as accuracy, loss, etc.) can be evaluated and visualized using matplotlib. Finally, we have trained our model. After the text-based binary classification model, we will try to use our deep learning model to predict heart disease and evaluate the results. We will now continue to use the text-based multi-class classification model. We will use the red wine quality data set from the Kaggle website. Our goal is to predict the multiple categories into which red wine samples can be placed based on the learning obtained from this data set. Repeat the same steps here.

The first step is to get the data set and load it into our program. Then, as the second step, we will perform EDA or exploratory data analysis on the uploaded data, and then we will prepare the data for our deep learning model. Then we will define the Keras deep learning model. Once we have defined the model, we will compile the model. Then we will fit our data set to the compiled model and wait for the training to complete. After training, training history and metrics (such as accuracy, loss, etc.) can be evaluated and visualized using matplotlib. Finally, we have trained our model. We will try to use the new data set to predict the quality of the wine, and then we will evaluate the classification results.

We may spend a lot of time, resources, and energy to train deep learning models. We will learn the technique of saving the trained model. This process is called serialization. We will first serialize a model. Then load it into another program and make predictions without repeating training. It is text-based data. We will now continue to process image-based data. In the preparatory course, we will introduce the basic concepts of digital imaging, in which we will understand the composition and structure of digital images, and then we will understand the use of Keras functions for basic image processing. There are many classes and functions in the Keras library API to help preprocess images.

We will learn about the most popular and useful features one by one. Another important and useful image processing function in Keras is an image enhancement, in which slightly different versions of images are automatically created during training. We will learn about single image enhancement, image enhancement within a directory structure, and data frame image enhancement. Then there is another theoretical session on the basis of convolutional neural networks or CNN.

We will learn how the basic CNN layers work, such as a convolutional layer, grouping layer, and fully connected layer. In image processing, there are concepts such as Stride Padding and Flattening in convolution. We will also learn one by one. Now we are ready to start using our CNN model. If images of flowers are provided in any of these categories, we will design a model that can classify 5 different types of flowers. We will first download the dataset from the Kaggle website. Then the first step will be to find this data set from our computer and load it into our program.

Then as the second step, we must manually split this data set to train and test the model. We organize them into training and testing folders, each class is marked in a separate folder, and then we will define the Keras deep learning model. Once we have defined the model, we will compile the model, then fit our data set on the compiled model and wait for the training to complete. After training, training history and metrics (such as accuracy, loss, etc.) can be evaluated and visualized using matplotlib. Finally, we have trained our model. We will try to use a new image dataset to predict five different types of flowers, and then we will evaluate the classification results.

We can use many techniques to improve the quality of the model. Especially image-based models. The most popular technique is the regularization of model loss. The next technique is the optimization and adjustment of the filters in the filling and convolutional layers, and finally the optimization through image magnification. We will modify different enhancement options in this meeting. It is a very tedious task to manually do these optimization techniques one by one and compare the results.

Therefore, we will use a technique called hyperparameter tuning, where the Keras library itself will change the different optimization techniques we specify and will report and compare the results without us interfering with it, although these techniques are interesting and creating models from scratch. If you plan to design a large model, this is time-consuming and may take several years. In this case, a technique called transfer learning can help us. We will carry the world’s abandonment, the most advanced technology, the majority of the population

## Who this course is for:

- Beginner who wants to learn the Basic to Advanced Deep Learning

Source: https://www.udemy.com/course/deep-learning-computer-vision-using-keras-dummies-guide/

Deep Learning using Keras – Complete & Compact Dummies Guide

**DOWNLOAD**

156 + Free courses Provided by Google Enroll Now

Coursera 1840 + Free Course Enroll Now

1500 + Free Online Courses of Udemy