Ā
Ā
Ā
Ā
Ā
AI Solutions Mastery
AI Practitioner Plus
Ā
Ā
Ā
Transition into a
Skilled End-to-End Data Scientist
(in 5 Weeks )
Ā
By Following a 10-Step Roadmap and Engaging in Comprehensive Hands-On Projects,
Including an Additional End-to-End Project to Solve Real-World Problems and Deploy Advanced AI Web Applications
Ā
Training is Suitable for you if:Ā
Individuals:
-
š¤Industry Experts:Ā You are an industry expert whose company wants to train in ML and AI to leverage your domain knowledge with advanced data science skills.
-
š¤Ā Professionals and PhD Graduates:Ā You are a professional (e.g., Finance Expert, IT Specialist) or a recent PhD graduate aiming to move into a data science role
-
š¤Ā Data Scientists:Ā You are a data scientist with limited real-life project experience seeking to apply best practices.
Companies:
-
š¤Ā Upskilling Internal Talent:Ā You are a company looking to upskill your internal talent in machine learning and AI to avoid the lengthy process of onboarding external data scientists who need to learn your business.
Possible Start Dates in 2024:Ā
- šĀ Fall:Ā 1 October | 1 NovemberĀ | 1 DecemberĀ
Reschedule :
- šĀ You can Reschedule to Any Month That Suits YouĀ
Immediate Access:Ā
- ā”Ā Get Immediate Access to Lessons Upon Registration
Certificate:Ā
- š Ā YesĀ
Duration:
- ā5-Weeks
- šTotal 14 hours Lessons
Ā
Price (excl. VAT):Ā Ā
ā¬1979
Ā
YOU WILLĀ LEARN
Ā
- Business Understanding:Ā Analyze real-world problems.
- Data Understanding & EDA:Ā Perform initial data cleaning and exploratory data analysis using Pandas, Matplotlib, Seaborn, and NetworkX.
- Data Preparation:Ā Prepare data for modeling.
- Feature Engineering:Ā Implement advanced feature engineering techniques.
- Modelling:Ā Build AI models (Supervised Learning: Build 9 different models including XGBoost; Deep Learning: TensorFlow, Keras; Unsupervised Learning: Anomaly Detection - Isolation Forest, AutoEncoders; Clustering - K-means, DBSCAN, Hierarchical Clustering).
- Model Fine-Tuning:Ā Enhance models with hyperparameter tuning using Hyperopt.
- Evaluation / Comparison:Ā Compare multiple models using different classification metrics such as AUCPR to select the best performer.
- AI Fairness:Ā Explore different types of parity to ensure fairness in AI models (FairLearn).
- AI Explainability:Ā Understand model decisions (SHAP).
- Deployment:Ā Develop and deploy models using web applications (Streamlit).
YOUR PROGRAM
Ā
Ā WEEK 1
Ā Ā Start the Engine (Start of 1st End to End Project)
-
- šĀ Get Started
- š„Ā Introduction to AI vs Machine Learning vs Deep Learning vs DS
- š„Ā Introduction to 10 Essential Steps of Data Science (Data Science Framework to 10X Your Performance)
- š„Ā Introduction to Business Understanding - Problem Understanding & Getting the Big Picture
- š„Ā Working With Real-World Data (HANDS-ON)
- š„Ā Data Understanding - Part 1: Setup and Data in Google ColabĀ (HANDS-ON)
- š„Ā Data Understanding - Part 2: Collect and Describe DataĀ (HANDS-ON)
- š„Ā Data Understanding - Part 3: Explore and Verify DataĀ (HANDS-ON)
- šĀ Assignment Week 1 - Apply Your Skills in Data Understanding & EDAĀ (HANDS-ON)
- āļø Personalized feedback on Weekly Assignments
Ā
Ā WEEK 2
Ā Ā AI SolutionĀ Engineered
-
- š„Ā Introduction to Week 2 - Data Prep - Feature Engineering - Modelling
- š„Ā Why Data Preparation and Feature Engineering
- š„Ā Why Modelling? Which Models? Model Evaluation Method
- š„Ā ScikitLearn Library - the Golden SourceĀ Ā (HANDS-ON)
- š„Ā Data Preparation: Setup Unique IDs and Stratified Test SetĀ Ā (HANDS-ON)
- š„Ā Data Preparation: Feature TransformationĀ Ā (HANDS-ON)
- š„Ā Feature EngineeringĀ Ā (HANDS-ON)
- š„Ā ModellingĀ Ā (HANDS-ON)
- šĀ Assignment Week 2: Apply Your Skills in Data Preparation, Feature Engineering, and ModelingĀ Ā (HANDS-ON)
- š¬Ā Q&A Live Session
- āļø Personalized feedback on Weekly Assignments
Ā
Ā WEEK 3
Ā Ā AI SolutionĀ Advance
-
- š„Ā Modelling - Supervised Learning - Classification - Regression
- š„Ā Introduction to Classification Metrics
- š„Ā ML Algorithms - Naive Bayes - Logistic Regression
- š„Ā ML Algorithms - Tree - Ensemble - Gradient Descent - RandomForest - Gradient Boosting
- š„Ā Solutions to Imbalanced Data - Part I - SMOTE - ADASYN
- š„Ā Introduction to Deep Learning and Its Concepts
- š„Ā Solutions to Imbalanced Data - Part II - GANs and Oversampling with CTGANs
- š„Ā Hyperparameter Tuning
- š„Ā Hyperparameter Tuning Using Bayesian and Tree-structured Parzen Estimators
- š„Ā XGBoost Hyperparameters
- š„Ā Supervised Classification XGBoost Deep Learning Hyperparameter Tuning CTGANsĀ (HANDS-ON)
- šĀ Assignment Week 3: Ensemble Classification - Deep Learning Oversampling with SMOTE, ADASYN, and CTGANs, and Hyperparameter Tuning
- āļø Personalized feedback on Weekly Assignments
Ā
Ā WEEK 4
Ā Ā AI SolutionĀ Ultimate
-
- š„Ā Introduction to ML Algorithms - Distance Similarity - KNN - Clustering - Anomaly Detection
- š„Ā Introduction to AI Explainability - Global vs Local Explainability - SHAP Values
- š„Ā Introduction to AI Fairness for Classification with Tabular Data - FairLearn Library
- š„Ā Final Model Selection - Deep Learning Ensemble Model Selection Hyperparameter TuningĀ (HANDS-ON)
- š„Ā Model Selection - Unsupervised Learning Anomaly Detection Dimensionality ReductionĀ (HANDS-ON)
- š„Ā AI Explainability - Global vs Local Explainability - SHAP ValuesĀ (HANDS-ON)
- š„Ā Hands-on AI Fairness with FairLearnĀ (HANDS-ON)
- šĀ Instruction to Install StreamlitĀ (HANDS-ON)
- š„Ā Workshop on Streamlit - Web Application Library for DeploymentĀ (HANDS-ON)
- š„Ā Complete Pipeline and Deployment with StreamlitĀ (HANDS-ON)
- šĀ Assignment Week 4 - Unsupervised Learning, Deep Learning, Explainability, Fairness, Pipeline and DeploymentĀ (HANDS-ON)
-
š¬Ā Q&A Live Session
- āļø Personalized feedback on Weekly Assignments
Ā
Ā WEEK 5
Ā
Ā Ā AI SolutionĀ Plus (2ndĀ End-to-End AIĀ Project)
- Overview: Apply the skillsĀ to a real-world problem from start to finish. This week, you will either select your own problem or choose one recommended by the course, prepare the data, build and optimize a model, evaluate it, and deploy it as a functioning AI product.
- āļø Personalized feedback on End-to-End Project
TRAINING ISĀ FOR YOUĀ IF:
Ā
ā You have some basic experience with Python.
āĀ Your Profile is any of the following:
- š¤ You are anĀ Industry ExpertĀ whose company wantsĀ to train in ML and AIĀ to leverage your domain knowledge with advanced data science skills.
- š¤ You are aĀ Professional (e.g., Finance Expert, IT Specialist)Ā or a recentĀ PhD graduateĀ aiming to move into a data science role.
- š¤ You are aĀ Data Scientist with Limited Real-life Project ExperienceĀ seeking to apply best practices.
āĀ You want to get certified in data science.
ā You want to learn from an AI expert with over 20 years of real-world experience.
āĀ You want to gain the skills to successfully apply data science in any field.
Ā
Ā
TRAININGĀ ISĀ NOT FORĀ YOUĀ IF:
Ā
ā Youāve never worked with Python.
ā You are already an experienced data scientist.
Ā
YOUR TRAINER
Dr. Maryam Miradi
WithĀ 20+ YearsĀ in AI development, a PhD in AI, 24 publications, 2 books, 5 awards, and experience coaching 200+ data scientists in 12 Industries,Ā Iām here to share AI Solutions with you.