
Explore dictionaries and sets in Python, mapping unique keys to values for fast lookup, and using sets to remove duplicates and perform intersection, union, and difference, including the get method.
Master Python comprehensions to build lists, dictionaries, sets, and generators in one line, using for loops, if conditions, and ternaries for filtering and transforming data.
Master robust error handling in Python using try-except blocks, else, and finally for cleanups. Catch specific errors like value error and zero division error, raise custom exceptions, and prevent crashes.
Explore object-oriented programming by building classes and objects, defining attributes and methods, and using constructors and self to manage state, with real examples of stock and account classes.
Navigate nested data in Python using json to map dicts and lists, and iterate records. Use json.dumps/loads for strings and json.dump/load for files.
Explore NumPy as the foundation of the Python data science stack, mastering fast n-dimensional arrays, vectorization, indexing, slicing, boolean masks, zeros, arrange, linspace, and sum, max, and mean aggregations.
Explore type hints and data classes to make Python code clearer, safer, and more self-documenting. See how hints enable static tooling and how dataclasses auto-generate boilerplate like constructors and representations.
Demonstrate Q-learning for an unmanned ground vehicle with NumPy, tuning alpha and gamma, building a 12x12 zero matrix, and mapping state to location for temporal-difference updates.
Master robust retrieval for RAG systems by combining hybrid search, re-ranking, and metadata filters to achieve accurate, fast, and safe finance document retrieval.
Compare fine-tuning, retrieval-augmented generation, and prompting to tailor language models. Learn a practical ladder from prompting to fine-tuning, emphasizing costs, use cases, and when to choose each.
Evaluate LLM systems by treating prompts as code, define an eval set as the specification, and use CI-driven regression checks with task-appropriate metrics to ensure reliable, drift-free production.
Master prompt engineering by turning prompts into precise instructions that unlock expert output from AI models, using roll, task, context, and constraints with temperature, top P, and tokens.
Identify and mitigate risks in deploying language models by implementing guardrails. Separate trusted data from instructions, ground answers, and red team against prompt injections, jailbreaks, and hallucinations.
Explore 500 ready-to-use prompt templates spanning foundations to production, organized into 12 categories for modern prompt engineering, with a bonus set of 60 power-user templates.
Machine Learning, Deep Learning & Time Series with Python is a comprehensive, practical course designed to take you from Python programming fundamentals to Machine Learning, Deep Learning, Reinforcement Learning, Time Series, Generative AI, Prompt Engineering, and real-world analytics applications.
The course begins with a strong Python foundation, covering variables, data types, operators, strings, lists, tuples, sets, dictionaries, conditionals, loops, functions, recursion, modules, comprehensions, file handling, exception handling, object-oriented programming, iterators, generators, decorators, regular expressions, dates and times, JSON, debugging, logging, testing, concurrency, performance profiling, SQL, web scraping, CLI tools, context managers, and Excel/CSV operations.
You will then work with important Python libraries for Machine Learning and Deep Learning, including NumPy, Pandas, Scikit-learn, TensorFlow, PyTorch, Keras, and Hugging Face.
The Machine Learning section covers Linear Regression, Logistic Regression, KNN, SVM, Decision Trees, Random Forests, Naïve Bayes, Ensemble Learning, K-Means, Hierarchical Clustering, DBSCAN, PCA, Association Rules, NLP, XGBoost, LightGBM, FastAI, JAX, and Q-Learning.
The course also includes a complete Reinforcement Learning project based on an Unmanned Ground Vehicle path-finding problem, covering the Bellman Equation, Markov Decision Process, Q-Learning, implementation, and path visualization.
In Deep Learning, you will learn Neural Networks, MLP, CNN, RNN, and LSTM. The Time Series section covers Time Series Foundations, ARIMA, and SARIMA.
The course goes beyond traditional Machine Learning with a dedicated AI & LLMs for Finance section covering LLM fundamentals, prompt engineering, structured outputs, embeddings, semantic search, RAG, advanced RAG, financial document intelligence, NLP-to-SQL, sentiment analytics, automated reporting, AI agents, workflow automation, analyst copilots, model selection, evaluation, security, and model risk governance.
You will also learn Prompt Engineering, including reasoning models, context engineering, structured outputs, tool use, agentic prompting, evaluation, optimization, safety, guardrails, and LLMOps.
The course further demonstrates how these analytical skills can be applied to real-world domains through dedicated projects and modules in Marketing Analytics, Actuarial Analytics, Credit Risk Analysis, and Financial Crime Analytics. Topics include customer segmentation, customer lifetime value, A/B testing, marketing mix modeling, actuarial risk, credit scorecards, WOE and Information Value, PD, LGD, EAD, IFRS 9, stress testing, fraud detection, anomaly detection, graph analytics, sanctions screening, entity resolution, and model explainability.
Additional Data Science content provides further practical coverage of Python, NumPy, SciPy, Pandas, Matplotlib, Seaborn, Plotly, data visualization, data cleaning, case studies, Machine Learning fundamentals, Scikit-learn, and NLP.
Whether you are starting your journey in Python and Data Science or want to expand into Machine Learning, Deep Learning, AI, LLMs, Time Series, and domain-specific analytics, this course provides a broad practical learning path combining programming, algorithms, projects, and real-world applications.