Convolutional Neural Networks: Deep Learning

Name: Convolutional Neural Networks: Deep Learning
Rating: 4.2 (25 reviews)

Gain a comprehensive understanding of CNNs and apply this knowledge to develop a project

Created bySujithkumar MA

Last updated 10/2023

English

What you'll learn

Understand the basics and types of 2D Signals (Images)
Understand and implement the process of convolution
Learn and implement the Convolutional neural networks for any real time applications
Review the fundamentals of deep learning

Course content

6 sections • 26 lectures • 3h 4m total length

Where does CNN lie?5:32
Explanation, Types of Deep Learning Networks7:29
Review deep learning fundamentals, from neural networks and weights to learning and memory in recurrent models like RNNs and LSTMs, and apply these to convolutional networks for image processing.

Convolution - A Deep Dive11:00
1D Convolution Example - HPF, LPF12:16
Basics of Images15:40
View images as two-dimensional signals of pixel intensities in binary, grayscale, and color formats. See RGB channels and eight-bit values (0–255), and how CNNs learn filters to detect edges.
Example of 2D Convolution12:41
Convolution in Action13:25
See convolution in action by applying a 2D kernel to a binary image in Python with OpenCV, NumPy, and Matplotlib, producing an output feature map.
Edge Detector Algorithm7:36
Mathematical model of CNN9:34
Explains convolution dimension rules in CNNs, showing how input size, filter size, stride, and padding determine output width via (w_in - w_f + 2p)/stride + 1, with integers.

Why CNN is ideal for Image Processing5:48
Discover why convolutional neural networks excel at image processing by automatically extracting and learning features with convolutional layers and fully connected layers, enabling adaptive filters and efficient image understanding.
Applications of CNN6:58
CNN Architecture and Layers2:25
Explore the cnn architecture, featuring feature learning and classification through convolutional layers with activation and pooling, followed by flattening, fully connected layers, and softmax output.
ReLU Activation0:03
How to perform Pooling9:02
Apply max pooling or average pooling to downsample feature maps after convolution, summarizing regions into single values. Flatten the pooled maps to feed fully connected layers for learning.
Batch Normalization16:10
Accelerate and stabilize learning in convolutional and fully connected networks with batch normalization, converting activations to zero mean and unit variance using batch mean and standard deviation plus epsilon.

Importing all the essential libraries5:18
Visualization and Preprocessing of Images10:04
Learn to download and organize the fruits 360 dataset from Kaggle into a Kaggle folder in Google Drive, then mount it in Google Colab and create training and testing folders.
Using glob to find out the number of classes1:22
Defining Convolutional Layers9:32
Training the CNN Architecture1:45
Train a convolutional neural network with an image data generator, fitting on train data for 20 epochs and using a validation generator to monitor accuracy and loss, reaching 83% accuracy.
The Testing Phase1:57

Requirements

No prior experience is needed. But a little bit of Python and Machine learning knowledge will make you more comfortable

Description

In this course, you'll be learning the fundamentals of deep neural networks and CNN in depth.

This course offers an extensive exploration of deep neural networks with a focus on Convolutional Neural Networks (CNNs).
The course begins by delving into the fundamental concepts to provide a strong foundation for learners.
Initial sections of the course include:
- Understanding what deep learning is and its significance in modern machine learning.
- Exploring the intricacies of neural networks, the building blocks of deep learning.
- Discovering where CNNs fit into the larger landscape of machine learning techniques.
- In-depth examination of the fundamentals of Perceptron Networks.
- Comprehensive exploration of Multilayer Perceptrons (MLPs).
- A detailed look into the mathematics behind feed forward networks.
- Understanding the significance of activation functions in neural networks.
A major portion of the course is dedicated to Convolutional Neural Networks (CNNs):
- Exploring the architecture of CNNs.
- Investigating their applications, especially in image processing and computer vision.
- Understanding convolutional layers that extract relevant features from input data.
- Delving into pooling layers, which reduce spatial dimensions while retaining essential information.
- Examining fully connected layers for making predictions and decisions.
- Learning about design choices and hyperparameters influencing CNN performance.
The course also covers training and optimization of CNNs:
- Understanding loss functions and their role in training.
- Grasping the concept of backpropagation.
- Learning techniques to prevent overfitting.
- Introduction to optimization algorithms for fine-tuning CNNs.
Practical implementation is a significant component:
- Hands-on coding and implementation using Python and deep learning frameworks like TensorFlow or PyTorch.
- Building and training CNN models for various applications.
- Gaining real-world skills to develop your own CNN-based projects.
By the course's conclusion, you'll have comprehensive knowledge of CNNs and practical skills for their application in various real-world scenarios. This knowledge empowers you in the field of deep learning and CNNs, whether you're interested in image recognition, object detection, or other computer vision tasks.

The last section is all about doing a project by implementing CNN

Who this course is for:

Anyone who wants to understand CNN in depth

Convolutional Neural Networks: Deep Learning

What you'll learn

Explore related topics

Course content

Introduction2 lectures • 13min

Neural Networks - A review4 lectures • 19min

Convolution in Digital Image Processing7 lectures • 1hr 22min

CNN - Layerwise study6 lectures • 40min

The Project - Fruits Classifier using CNN6 lectures • 30min

Resources1 lecture • 1min

Requirements

Description

Who this course is for: