Module 1:

Data collection

Introduction to data pipelines
Introduction to APIs
HTTP requests
Selenium for web scraping
Performing basic web automation actions
Referencing HTML elements using Xpath
Cron

Module 2:

Data formats & the Pandas Python library

CSV, JSON & Parquet file formats
Intro to Pandas

Module 3:

Data cleaning

Reasons for and approaches to data cleaning
Handling missing data

Module 4:

Intro to cloud

Data Lakes and Warehouses
AWS S3
AWS DynamoDB
AWS RDS

Module 5:

SQL

SQL basics
SQL Join operations
SQL Aggregations
SQL Subqueries

Module 6:

ETL, Distributed Computing & Data Versioning

Intro to DataBricks
Distributed computing with PySpark
ETL
Data Versioning

Module 1:

Data science fundamentals

Statistical modelling
Hypothesis testing
Multi-collinearity
Influential points

Module 2:

Introduction to machine learning

Intro to ML & when (NOT) to use
Scikit-learn
Validation and testing
Linear regression
Logistic regression
Multiclass classification
Hyperparameters, grid search and K-fold cross validation
Feature selection and feature engineering

Module 3:

Theory

Evaluation metrics
Bias & variance, underfitting and overfittting
Regularisation
Maximum Likelihood Estimation (MLE)

Module 4:

More supervised machine learning models

K-nearest neighbours
Decision trees
Random forests
Gradient boosting

Module 5:

Unsupervised machine learning models

K-means clustering
Dimensionality reduction using principal component analysis (PCA) & T-SNE
DBSCAN

Module 1:

PyTorch

Automatic differentiation
PyTorch Datasets and DataLoaders
Making custom datasets

Module 2:

Neural Networks

Neural networks
Dropout
Batch Normalisation
Optimisation for deep learning
Convolutional Neural Networks (CNNs)
ResNets

Module 3:

Practical

Architecture tips, data augmentation & debugging tips
Pre-trained models
Transfer learning
Hardware acceleration (GPUs & TPUs)

Module 4:

Advanced Architectures

Attention
Self-attention
Transformers

Module 5:

Applications

Churn modelling
Content based recommendation systems
Collaborative filtering
Lead scoring

Module 1:

Scaling experiments in the cloud

EC2, AMIs & SSH
Experiment tracking
Ray Tune

Module 2:

Managed Solutions

AWS Sagemaker
Architecting an end-to-end ML product on AWS

Module 3:

Intro to MLOps

MLOps overview
ETL
Feature stores
Data versioning
Orchestration & Airflow

Module 4:

Custom Solutions

Docker
Flask
MLFlow
TorchServe
Prometheus
Grafana

Module 5:

Fault-tolerant deployment and auto-scaling

CI/CD with GitHub actions
Kubernetes
AWS EKS
Kubeflow
AWS Cloudformation