Keras: Practical AI Projects & Deep Learning using Keras

Name: Keras: Practical AI Projects & Deep Learning using Keras
Rating: 4.3 (3 reviews)

Explore practical AI projects, including chatbots, sentiment analysis, image classification, advanced face recognition

Created byEDUCBA Bridging the Gap

Last updated 3/2024

English

What you'll learn

Building chatbots using Keras. Sentiment analysis implementation with recurrent neural networks (RNN).
Image classification techniques using Keras. Advanced face recognition applications using computer vision and deep learning.
Practical project implementation on Google Colab. Text preprocessing techniques like Bow Model, Count Vectorizer, Stemming, and Lemmatization.
Model training, evaluation, and prediction. Pretrained model utilization and fine-tuning. Image preprocessing, augmentation, and visualization.
Face detection and recognition algorithms. Embedding generation and classification. Real-world implementation and testing of AI models.

Course content

4 sections • 73 lectures • 9h 26m total length

Introduction to Project5:48
Bow Model8:08
Master text data handling for chatbots by preprocessing sentiment data and applying bag-of-words models, including count vector, term frequency, tf-idf, and glow model.
Count Vectorizer10:46
Text Data7:34
Text Data Continue9:47
Limit Number of Features8:16
Stop Words7:49
Stemming10:32
Stemming Continue10:07
Lemmatization6:41
ML Model on Text Data8:24
TF-TF-IDF Vectorizer5:30
Spacy Word2Vec8:43
Learn to measure word similarity with SpaCy using a pretrained word-to-vector model. Install SpaCy, download a language model, load it, and compute token similarities to build a basic chatbot.
Requirements6:54
Hindson Implementation6:41
Hindson Implementation Continue9:13
Neural Networks9:16
Generative Chatbots Part 19:33
Generative Chatbots Part 26:58
Generative Chatbots Part 312:22
Generative Chatbots Part 48:02
Generative Chatbots Part 56:02
Generate text with an LSTM-based generative chatbot by predicting the next character from a seed, using softmax probabilities and argmax to select the highest value.
Attentive Chatbots Part 111:08
Attentive Chatbots Part 25:56
Attentive Chatbots Part 35:20
Build an attentive chatbot by creating a subword tokenizer from a questions-and-answers corpus using tfds, and define start and end tokens; test tokenization and pad sequences with tf.keras.
Advanced Chatbot10:55
Advanced Chatbot - Evaluation2:57
Evaluate a chatbot by generating predictions for test data, comparing outputs, and tracking accuracy (16%), then improve with more epochs and production environment tips.
Conclusion5:42

Introduction to Project11:18
Develop a binary image classification project in Keras using a pre-trained VGG16 model from ImageNet, with CNN layers, max pooling, image preprocessing and augmentation, and transfer learning.
Google Collab3:43
Uploading7:27
Downloading the Dataset7:06
Pretrained Model5:01
Intermediate Layer Visualization6:08
Model Creation and Image Augmentation9:20
Compiling and Training Model7:42
Extend the VGG16 base with added fully connected layers and compile and train a binary cat-vs-dog classifier using flow from directory with image augmentation, rescaling, rmsprop 0.001, and early stopping.
Loss Values7:10
Test Images and Visualization8:12
Evaluate test performance with a Keras test generator reading 224×224 images for a binary classifier, report about 49% accuracy on 1000 test images, and visualize sample predictions.
Retraining the Model7:24

Introduction to Course6:11
CNN for Image Processing10:38
Image Preprocessing10:21
Learn how to preprocess image data for Keras models by reading and decoding image files to RGB, converting to floats, and rescaling pixel values using the Keras image data generator.
Saving and Loading the Models6:00
Getting System Ready4:02
Reading the Image Data6:28
Detect Faces MTCNN9:20
Use a pre-trained empty cnn detector (mtcnn) to identify faces in an image, returning bounding boxes and facial key points for each face; the example detects three faces.
Draw Bounding Box7:55
Draw Key points8:18
Apply on Group of Images5:48
Extract Faces from Image7:31
Face Detection Summary2:39
Face Recognition8:46
Fashion Dataset11:23
Load Faces6:43
Load Dataset from Folders8:31
Load Dataset from Folders Continue5:59
Generate Face Embeddings12:04
Face Embeddings5:04
Building Classifier on Embeddings6:50
Load saved face embeddings from a compressed npz file, normalize with an L2 normalizer for unit-length vectors used in euclidean distance classification, and prepare train and validation data.
Building Classifier on Embeddings Continue7:21
Testing for Real Implementation11:09
Use Kera's DNN with Face net11:12
Integrate pre-trained FaceNet into a Keras sequential model, freeze its weights, add dense layers, and train with image data generators on 160 by 160 inputs for five-class softmax output.
Conclusion9:32

Requirements

Python programming language.
Fundamentals of machine learning and deep learning concepts.

Description

Welcome to the comprehensive course on practical applications of deep learning with Keras! In this course, you will embark on an exciting journey through various projects aimed at developing practical skills in deep learning and neural networks using the Keras framework. Whether you're a beginner looking to get started with deep learning or an experienced practitioner seeking to enhance your skills, this course offers something for everyone.

Throughout this course, you will dive into hands-on projects covering a wide range of topics, including building chatbots, sentiment analysis using recurrent neural networks (RNNs), image classification, and advanced face recognition computer vision applications. Each project is carefully designed to provide you with practical experience and insights into real-world applications of deep learning.

By the end of this course, you will have gained valuable experience in implementing deep learning models, understanding their underlying principles, and applying them to solve complex tasks. Whether you're interested in natural language processing, computer vision, or any other domain, the skills you acquire in this course will be invaluable in your journey as a deep learning practitioner.

Get ready to unlock the full potential of deep learning with Keras and take your skills to the next level!

Section 1: Building A Chatbot with keras

In this section, students will embark on a practical journey of constructing a chatbot using Keras. They will begin with an introduction to the project's objectives, followed by an exploration of foundational concepts such as the Bag of Words (BoW) model, Count Vectorizer, and techniques for handling text data. Through a series of progressive lectures, students will delve into preprocessing steps, feature limitation strategies, and essential text processing elements like stop words and stemming.

Section 2: Project On Keras: Sentimental Analysis Using RNN

In the second section, students will transition to another project focusing on sentiment analysis with Recurrent Neural Networks (RNNs) using Keras. They will be introduced to Google Colab for collaborative work and IMBD dataset for sentiment analysis. The section will cover topics such as padding sequences, basic and complex LSTM models, and training procedures, enabling students to gain practical experience in sentiment analysis.

Section 3: Project On Keras - Image Classification

Continuing the journey, students will move to image classification projects in this section. They will learn to set up Google Colab, download datasets, and employ pretrained models for image classification tasks. Topics covered will include intermediate layer visualization, model creation, image augmentation, and model evaluation techniques.

Section 4: Project On Keras - Creating An Advanced Face Recognition Computer Vision App

In the final section, students will engage in creating an advanced face recognition application using computer vision techniques with Keras. They will explore Convolutional Neural Networks (CNNs) for image processing, face detection using MTCNN, and building a classifier for face recognition. This section will culminate in a comprehensive understanding of implementing deep learning models for real-world applications.

Who this course is for:

Students or professionals seeking to enhance their skills in machine learning and deep learning.
Data scientists looking to expand their knowledge in natural language processing (NLP) and computer vision.
Software engineers interested in developing advanced applications using Keras and TensorFlow.
Individuals aspiring to build chatbots, perform sentiment analysis, and work on image classification and face recognition projects.
Professionals seeking to advance their careers in artificial intelligence (AI) and deep learning-related roles.
Anyone with a keen interest in exploring advanced projects in the field of artificial intelligence and machine learning.

Keras: Practical AI Projects & Deep Learning using Keras

What you'll learn

Explore related topics

Course content

Building A Chatbot with keras28 lectures • 3hr 45min

Project On Keras: Sentimental Analysis Using RNN10 lectures • 1hr 11min

Project On Keras - Image Classification11 lectures • 1hr 21min

Project On Keras - Creating An Advanced Face Recognition Computer Vision App24 lectures • 3hr 10min

Requirements

Description

Who this course is for: