Deep Learning – Crash Course 2023

Packt via Coursera

Coursera based on 0 ratings

Likes

Overview

Unlock the power of deep learning and elevate your machine learning skills with our comprehensive deep neural networks course. This hands-on program covers deep learning fundamentals, including artificial neural networks, activation functions, bias, data, and loss functions. Learn Python basics focused on data science, and master tools like Matplotlib, Num Py, and Pandas for data cleaning and visualization. Progress from the MP Neuron model to the Perceptron, Sigmoid Neuron, and Universal Approximation Theorem, exploring Re LU and Soft Max activation...

Syllabus

Welcome on Board

In this module, we will welcome you to the course and provide an overview of deep learning. We will explain the course objectives, the structure of the content, and the skills and knowledge you will acquire throughout the course.

Getting the Basics Right

In this module, we will lay the foundation for understanding deep learning by covering essential topics such as artificial neural networks, activation functions, and bias. We will also explore the role of data, various applications, models, loss functions, and learning algorithms crucial for model performance.

Python Crash Course on Basics

In this module, we will provide a crash course on the basics of Python programming, essential for deep learning. You will learn how to install and use Jupyter Notebook and Google Colab, understand data types, containers, control statements, and implement functions and classes in Python.

Python for Data Science - Crash Course

In this module, we will delve into Python libraries crucial for data science. You will learn how to handle arrays with NumPy, manipulate data using Pandas, and visualize data with Matplotlib. We will cover topics from basic data structures to advanced data cleaning and plotting techniques.

MP Neuron Model

In this module, we will explore the MP Neuron model, also known as the McCulloch-Pitts model. You will gain an understanding of the data intuition, learn how to find parameters, and develop a mathematical intuition for this fundamental concept in neural networks.

MP Neuron in Python

In this module, we will focus on implementing the MP Neuron model in Python. You will learn how to import datasets, apply train-test split, and modify data. By the end of this section, you will have created an MP Neuron class from scratch and practiced with an assignment.

Summary of MP Neuron

In this module, we will summarize the key concepts and practical implementation of the MP Neuron model. We will review the important points and ensure you have a solid understanding through a recap and evaluation assignments.

Perceptron

In this module, we will cover the Perceptron model, discussing its representation, loss function, and parameter updates. You will understand how the update rule works and see its practical implementation in programs.

Perceptron in Python

In this module, we will implement the Perceptron model in Python. You will learn to program the model and visualize its accuracy and performance with increasing epochs, enhancing your practical skills in deep learning.

Sigmoid Neuron

In this module, we will transition from Perceptron to Sigmoid Neuron. You will learn about the limitations of the Perceptron, the benefits of the Sigmoid Neuron, and gain insights into gradient descent for model optimization.

Sigmoid Neuron Implement with Python

In this module, we will implement the Sigmoid Neuron using Python. You will learn to download and standardize datasets, and create a class for the Sigmoid activation function, solidifying your understanding through practical assignments.

Basic Probability

In this module, we will cover basic probability concepts. You will learn about random variables, their importance, types, and probability distribution tables, as well as the concept of entropy loss in the context of deep learning.

Deep Neural Networks

In this module, we will explore deep neural networks. You will learn why they are important, and through practical programming, understand the concept of linear separation of data, preparing you for more complex deep learning models.

Universal Approximation Theorem

In this module, we will delve into the Universal Approximation Theorem. You will learn its significance, confirm its effectiveness with practical examples, and discuss the challenges of building deep neural networks from scratch.

Deep Learning with TensorFlow 2.x

In this module, we will focus on TensorFlow 2.x for deep learning. You will learn to build, train, and evaluate neural networks using TensorFlow, with a recap of deep learning concepts and a summary to prepare for more advanced topics.

Activation Functions in Deep Learning Neural Networks

In this module, we will cover activation functions in deep learning. You will learn about different activation functions provided by TensorFlow and understand common network configurations used in deep learning tasks.

Applying Deep Learning

In this module, we will apply deep learning concepts. You will transition from shallow to deep learning, understand Keras basics, solve classification and regression problems, and explore advanced TensorFlow techniques and subclassing methods.