Learn Machine learning & AI [Beginner to advanced content]

Name: Learn Machine learning & AI [Beginner to advanced content]
Rating: 4.1 (1279 reviews)

Get well versed with Machine learning and AI by working on Hands-on Projects.

Created bywww.edyoda.com EdYoda

Last updated 7/2024

English

English [Auto],

What you'll learn

Machine Learning and its applications
Building End to End Machine Learning Projects
Deployment of Machine & Deep learning algorithms

Course content

3 sections • 34 lectures • 1h 57m total length

Project Requirements & Django Project Creation7:56
Develop and deploy a Django web app for image upload and automatic prediction using a deep learning model, including project setup, templates, views, and basic routing.
Backend creation using pretrained Keras model - Resnet506:34
Adding Form to django App & uploading image5:35
Learn to add a form to a Django app, handle post requests with multipart data, and upload and save images. Connect forms with views to process uploads and display results.
Integrating web application with deep learning backend8:57
Connect a deep learning backend to a web app by importing libraries, creating the model, preprocessing input, and rendering image predictions with confidence scores.

AI Project : Fraud Detection -Introduction0:57
Explore fraud detection by building production-quality prediction models through five modules, covering objectives, infrastructure, installation, data preprocessing, algorithm comparison, selection, and deployment on a local machine.
Module 1 : Agenda0:34
Discover the course objective and what constitutes credit card fraud, and learn how Kafka and Cassandra will be used to build an integrated solution.
Module 1 : Understanding Objective1:27
Analyze the data flow of the internet ecosystem and how historical credit card convictions train a machine learning model to predict fraud, deploy it, and retrain with new data.
Module 1: Kafka for gathering live credit card data1:23
Explore how Kafka, a scalable distributed streaming platform, uses a publish-subscribe model where producers submit transaction data to topics and a model consumer makes predictions on those credit card transactions.
Module 1 : Cassandra for storing data0:48
Cassandra stores historical and life prediction data, enabling high-volume writes at massive scale. The caption notes assessing data streaming source and storing data at various levels, with a proposed solution.
Module 1: Solution System Design2:02
Design a machine learning solution by integrating Cassandre data, scalable training with Weisbach, and a Flask-based REST deployment, enabling online and offline learning workflows.
Module 2: Agenda0:32
Explore the module 2 agenda for credit card fraud detection, detailing system requirements, package installation, and Cassandre installation steps.
Module 2: System Requirements0:43
Explore system requirements for setting up a local machine for machine learning, compare Windows dependencies, and consider using virtual machines or dual-boot setups for smoother operation.
Module 2: Java Installation0:27
Install java as required, and on ubuntu run commands one to four; for other systems use equivalent commands, then proceed to Spark installation.
Module 2: Kafka Installation2:18
Learn kafka installation by downloading binaries and configuring the package. Start zookeeper, then set up topics for producers and consumers.
Module 2: Anaconda Installation0:32
Discover the Anaconda package manager essential for machine learning, and learn how to download the correct version for Windows, Mac, or Linux.
Module 2: Docker Installation1:04
Discover docker installation as a container technology and learn how to install Cassandra using docker compose to run Cassandra with multiple nodes on a single machine.
Module 2: Cassandra Installation2:58
Install cassandra with docker, pull the image, run two containers with names, connect them, and access the sequel interface via the container IP address to manage nodes.
Module 3: Agenda0:42
Learn why preprocessing is required, how to use a third-party package, balance and sampling techniques, and how to choose estimates and build a pipeline in module 3.
Module 3: Loading & Preprocessing of Data1:45
Learn why preprocessing turns data into numerical features for machine learning and how to load a subset while separating features from the class target. Address class imbalance with sampling techniques.
Module 3 : Sampling Techniques3:03
Explore sampling techniques for imbalanced data using the imbalanced package, including oversampling and undersampling, and visualize how generated data reshapes the current class distribution.
Module 3: Estimators to be considered1:22
Module 3: Connecting transformers & estimators using Pipeline2:21
Apply sampling techniques to balance the data, then estimate it using a pipeline that combines models and samplers. Train, test, and validate multiple pipeline combinations before tuning.
Module 4: Agenda0:51
Explore code flow and component interactions in credit card fraud detection project; design a model class with training and validation parameters, train on subset, read/write to Cassandra, validate, and deploy.
Module 4: Code Flow1:50
Learn to convert a data schema into a class with three classes: a data schema class, a Cassandra read/write class, and a model class that holds training data and models.
Module 4: Class to connect Cassandra4:14
Learn how to configure a Cassandra client, map a data class, and read and write to a keyspace by syncing multiple sources with driver and persisting dictionaries to the database.
Module 4: Class code for Model Config & Training3:30
Module 4: Hyperparameter Tuning9:38
Module 4: Stitching all components4:03
Stitch together a complete machine learning solution by integrating data flow from Cassandra, configuring parameters, and testing multiple algorithmic permutations, while monitoring system usage during long runs.
Module 5: Agenda0:50
Evaluate multiple trained pipelines to select the best model, understand the architecture, and deploy by running the model in production, predicting with producer data, and serving behind the interface.
Module 5: Model Persistence & Selection2:02
Evaluate model performance using a confusion matrix, compare multiple models, and save the best one for persistence and deployment, validating all versions before final deployment.
Module 5: Revisiting Solution1:47
Revisit the designed solution, review Cassandre's write/read components, deploy model three, ensure producer data flows to the consumer with persistence, and prepare a Flask interface for predictions and COFCO integration.
Module 5 : Model behind Kafka Consumer6:23
Module 5: Model behind REST using Flask5:29
Learn to deploy a credit card fraud detection model behind a REST server using Flask, including loading the model, preprocessing data, predicting transactions, and writing results to a database.

Requirements

Python
scikit-learn
Machine learning
Deep Learning
Flask

Description

Do you feel overwhelmed going through all the AI and Machine learning study materials?

These Machine learning and AI projects will get you started with the implementation of a few very interesting projects from scratch.

The first one, a Web application for Object Identification will teach you to deploy a simple machine learning application.

The second one, Dog Breed Prediction will help you building & optimizing a model for dog breed prediction among 120 breeds of dogs. This is built using Deep Learning libraries.

Lastly, Credit Card Fraud detection is one of the most commonly used applications in the Finance Industry. We talk about it from development to deployment. Each of these projects will help you to learn practically.

Who's teaching you in this course?

I am Professional Trainer and consultant for Languages C, C++, Python, Java, Scala, Big Data Technologies - PySpark, Spark using Scala Machine Learning & Deep Learning- sci-kit-learn, TensorFlow, TFLearn, Keras, h2o and delivered at corporates like GE, SCIO Health Analytics, Impetus, IBM Bangalore & Hyderabad, Redbus, Schnider, JP Morgan - Singapore & HongKong, CISCO, Flipkart, MindTree, DataGenic, CTS - Chennai, HappiestMinds, Mphasis, Hexaware, Kabbage. I have shared my knowledge that will guide you to understand the holistic approach towards ML.

Here are a few reasons for you to pursue a career in Machine Learning:

1) Machine learning is a skill of the future – Despite the exponential growth in Machine Learning, the field faces skill shortage. If you can meet the demands of large companies by gaining expertise in Machine Learning, you will have a secure career in a technology that is on the rise.
2) Work on real challenges – Businesses in this digital age face a lot of issues that Machine learning promises to solve. As a Machine Learning Engineer, you will work on real-life challenges and develop solutions that have a deep impact on how businesses and people thrive. Needless to say, a job that allows you to work and solve real-world struggles gives high satisfaction.
3) Learn and grow – Since Machine Learning is on the boom, by entering into the field early on, you can witness trends firsthand and keep on increasing your relevance in the marketplace, thus augmenting your value to your employer.
4) An exponential career graph – All said and done, Machine learning is still in its nascent stage. And as the technology matures and advances, you will have the experience and expertise to follow an upward career graph and approach your ideal employers.
5) Build a lucrative career– The average salary of a Machine Learning engineer is one of the top reasons why Machine Learning seems a lucrative career to a lot of us. Since the industry is on the rise, this figure can be expected to grow further as the years pass by.
6) Side-step into data science – Machine learning skills help you expand avenues in your career. Machine Learning skills can endow you with two hats- the other of a data scientist. Become a hot resource by gaining expertise in both fields simultaneously and embark on an exciting journey filled with challenges, opportunities, and knowledge.

Machine learning is happening right now. So, you want to have an early bird advantage of toying with solutions and technologies that support it. This way, when the time comes, you will find your skills in much higher demand and will be able to secure a career path that’s always on the rise.

Practical Learning !!
Project-based learning has proven to be one of the most effective ways to engage students and provide a practical application for what they’re learning and it provides opportunities for students to collaborate or drive their learning, but it also teaches them skills such as problem-solving and helps to develop additional skills integral to their future, such as critical thinking and time management

By pursuing this course you will able to understand the concept of Machine learning at the next level you will also get to know about Artificial intelligence and that will boost your skill set to be a successful ML engineer.

Enroll now, see you in class!!

Happy learning!
Team Edyoda

Who this course is for:

Python developers looking for a future in machine learning and Artificial Intelligence (AI).
Machine learning learners looking for live projects to learn from.
Python and Machine learning developers looking for hands on projects to work on Artificial Intelligence (AI).

Learn Machine learning & AI [Beginner to advanced content]

What you'll learn

Explore related topics

Course content

Project 1 - AI Project Web application for Object Identification4 lectures • 29min

Project 2 - Deep Learning Project : Dog Breed Prediction (Kaggle dataset)1 lecture • 23min

AI Project : End to End Credit Card Fraud Detection29 lectures • 1hr 6min

Requirements

Description

Who this course is for: